From c24b34b6dbd245d6ab878124d8d5546fe9be96ab Mon Sep 17 00:00:00 2001 From: kowshik Date: Wed, 22 Mar 2023 21:09:45 +0000 Subject: [PATCH] Upload files to '' --- Decission_Tree_Wine_Recommendation.ipynb | 2555 ++++++++++++++++++++++ pipe.pkl | Bin 0 -> 6191 bytes wine_data.xlsx | Bin 0 -> 10264 bytes 3 files changed, 2555 insertions(+) create mode 100644 Decission_Tree_Wine_Recommendation.ipynb create mode 100644 pipe.pkl create mode 100644 wine_data.xlsx diff --git a/Decission_Tree_Wine_Recommendation.ipynb b/Decission_Tree_Wine_Recommendation.ipynb new file mode 100644 index 0000000..380ed1c --- /dev/null +++ b/Decission_Tree_Wine_Recommendation.ipynb @@ -0,0 +1,2555 @@ +{ + "nbformat": 4, + "nbformat_minor": 0, + "metadata": { + "colab": { + "provenance": [] + }, + "kernelspec": { + "name": "python3", + "display_name": "Python 3" + }, + "language_info": { + "name": "python" + } + }, + "cells": [ + { + "cell_type": "code", + "execution_count": 92, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "R9zvSxjt24af", + "outputId": "de9749be-551f-4613-e586-b70389d852e3" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Dimensions: 178 x 13\n", + "\n", + "Header: ['alcohol', 'malic acid', 'ash', 'ash alcalinity', 'magnesium', 'total phenols', 'flavanoids', 'nonflavanoid phenols', 'proanthocyanins', 'color intensity', 'hue', 'OD280/OD315 of diluted wines', 'proline']\n", + "1st row [1.423e+01 1.710e+00 2.430e+00 1.560e+01 1.270e+02 2.800e+00 3.060e+00\n", + " 2.800e-01 2.290e+00 5.640e+00 1.040e+00 3.920e+00 1.065e+03]\n" + ] + } + ], + "source": [ + "from mlxtend.data import wine_data\n", + "X, y = wine_data()\n", + "\n", + "print('Dimensions: %s x %s' % (X.shape[0], X.shape[1]))\n", + "print('\\nHeader: %s' % ['alcohol', 'malic acid', 'ash', 'ash alcalinity',\n", + " 'magnesium', 'total phenols', 'flavanoids',\n", + " 'nonflavanoid phenols', 'proanthocyanins',\n", + " 'color intensity', 'hue', 'OD280/OD315 of diluted wines',\n", + " 'proline'])\n", + "print('1st row', X[0])" + ] + }, + { + "cell_type": "code", + "source": [ + "import numpy as np\n", + "print('Classes: %s' % np.unique(y))\n", + "print('Class distribution: %s' % np.bincount(y))" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "kj6nE9hz3AMs", + "outputId": "ae7a436c-6bcd-43de-c171-c34034ef4320" + }, + "execution_count": 93, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Classes: [0 1 2]\n", + "Class distribution: [59 71 48]\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "X.shape" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "lQ_JL1i4Eo0C", + "outputId": "2971e185-5ee1-4785-e6ff-05d042406441" + }, + "execution_count": 94, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(178, 13)" + ] + }, + "metadata": {}, + "execution_count": 94 + } + ] + }, + { + "cell_type": "code", + "source": [ + "columns = ['alcohol', 'malic acid', 'ash', 'ash alcalinity',\n", + " 'magnesium', 'total phenols', 'flavanoids',\n", + " 'nonflavanoid phenols', 'proanthocyanins',\n", + " 'color intensity', 'hue', 'OD280/OD315 of diluted wines',\n", + " 'proline']" + ], + "metadata": { + "id": "LIyPC2bsEyEq" + }, + "execution_count": 95, + "outputs": [] + }, + { + "cell_type": "code", + "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=0.3, random_state=42)" + ], + "metadata": { + "id": "KUIpscN0I3T4" + }, + "execution_count": 96, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "from sklearn import tree\n", + "clf = tree.DecisionTreeClassifier()" + ], + "metadata": { + "id": "bmMYttf2M7GY" + }, + "execution_count": 97, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "# Fit the classifier to the training data\n", + "clf.fit(X_train, y_train)\n", + "\n", + "# Predict the class labels for the test data\n", + "y_pred = clf.predict(X_test)\n", + "\n", + "# Calculate the accuracy of the classifier\n", + "accuracy = clf.score(X_test, y_test)\n", + "\n", + "# Print the results\n", + "print(\"Predicted class labels:\", y_pred)\n", + "print(\"Accuracy:\", accuracy)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "6mGo2Xb3NX3X", + "outputId": "4a9d3837-1c80-41ed-fbfb-ad28c2d0c45b" + }, + "execution_count": 98, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Predicted class labels: [0 0 2 0 1 0 1 2 1 2 1 0 0 1 0 1 1 1 0 1 0 1 1 2 2 2 1 1 1 0 0 1 2 0 0 0 2\n", + " 2 1 2 1 1 1 1 2 0 1 1 2 0 1 0 0 2]\n", + "Accuracy: 0.9444444444444444\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "import matplotlib.pyplot as plt\n", + "plt.figure(figsize=(12,8))\n", + "tree.plot_tree(clf)\n", + "plt.show()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 466 + }, + "id": "UeqGcdxwN7CE", + "outputId": "bf7ad83c-399e-4127-c273-a83cb9296460" + }, + "execution_count": 99, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "# Confusion Metrics" + ], + "metadata": { + "id": "6ApROW4KO5Qw" + } + }, + { + "cell_type": "code", + "source": [ + "from sklearn.metrics import confusion_matrix\n", + "import seaborn as sns\n", + "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score\n", + "plt.figure(figsize=(12,8))\n", + "cm = confusion_matrix(y_test,y_pred)\n", + " \n", + "#Plot the confusion matrix.\n", + "sns.heatmap(cm,\n", + " annot=True,\n", + " fmt='g',\n", + " xticklabels=['0', '1' , '2'],\n", + " yticklabels=['0', '1' , '2'])\n", + "plt.ylabel('Prediction',fontsize=13)\n", + "plt.xlabel('Actual',fontsize=13)\n", + "plt.title('Confusion Matrix',fontsize=17)\n", + "plt.show()\n", + " \n", + " " + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 520 + }, + "id": "bCc29IiJOLKI", + "outputId": "b5754110-096a-4a5a-8f42-3a72af060178" + }, + "execution_count": 100, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + } + ] + }, + { + "cell_type": "code", + "source": [ + "from sklearn.metrics import accuracy_score\n", + "accuracy_score(y_test,y_pred)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "KGoGXwRkPHBH", + "outputId": "10bfeed7-0d53-409e-ad12-723de04671f6" + }, + "execution_count": 101, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "0.9444444444444444" + ] + }, + "metadata": {}, + "execution_count": 101 + } + ] + }, + { + "cell_type": "code", + "source": [ + "from sklearn.metrics import classification_report\n", + "print(classification_report(y_test, y_pred))" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "TOzbcfXSQUGo", + "outputId": "24336c98-a019-44a4-c55c-cdccd3a4da24" + }, + "execution_count": 102, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + " precision recall f1-score support\n", + "\n", + " 0 0.94 0.89 0.92 19\n", + " 1 0.91 1.00 0.95 21\n", + " 2 1.00 0.93 0.96 14\n", + "\n", + " accuracy 0.94 54\n", + " macro avg 0.95 0.94 0.95 54\n", + "weighted avg 0.95 0.94 0.94 54\n", + "\n" + ] + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "### Wine Classificaion using the wine_data.xlsx" + ], + "metadata": { + "id": "Lbhxu3u6i4at" + } + }, + { + "cell_type": "markdown", + "source": [ + "### Loading the data" + ], + "metadata": { + "id": "wCtbWMDejBW9" + } + }, + { + "cell_type": "code", + "source": [ + "wine = pd.read_excel('/content/wine_data.xlsx')" + ], + "metadata": { + "id": "9Udcqv9QQ1f_" + }, + "execution_count": 163, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "wine.head()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 206 + }, + "id": "bUDrNfnHjtBs", + "outputId": "d8d0ea56-ffd8-4c2c-ef49-22bc745161c4" + }, + "execution_count": 164, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " Preference Red_Wine White_Wine Recommendation\n", + "0 Red Light-Bodied NaN Pinot Noir\n", + "1 Red Full-Bodied NaN Shiraz or Zinfandel\n", + "2 White NaN Dry Sauvignon Blanc\n", + "3 White NaN Sweet Gewurztraminer\n", + "4 Red-Fruity NaN NaN Pinot Noir" + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
PreferenceRed_WineWhite_WineRecommendation
0RedLight-BodiedNaNPinot Noir
1RedFull-BodiedNaNShiraz or Zinfandel
2WhiteNaNDrySauvignon Blanc
3WhiteNaNSweetGewurztraminer
4Red-FruityNaNNaNPinot Noir
\n", + "
\n", + " \n", + " \n", + " \n", + "\n", + " \n", + "
\n", + "
\n", + " " + ] + }, + "metadata": {}, + "execution_count": 164 + } + ] + }, + { + "cell_type": "code", + "source": [ + "wine.value_counts()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "Nxoyk2DUjvwD", + "outputId": "f63bb36f-d3a3-4696-e128-2a6f53da402b" + }, + "execution_count": 165, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "Series([], dtype: int64)" + ] + }, + "metadata": {}, + "execution_count": 165 + } + ] + }, + { + "cell_type": "code", + "source": [ + "wine.info()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "XM0ZWZq_j1Dj", + "outputId": "c4f3eaa0-1a30-4806-fb3a-81329552fb82" + }, + "execution_count": 166, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "\n", + "RangeIndex: 40 entries, 0 to 39\n", + "Data columns (total 4 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 Preference 40 non-null object\n", + " 1 Red_Wine 18 non-null object\n", + " 2 White_Wine 6 non-null object\n", + " 3 Recommendation 35 non-null object\n", + "dtypes: object(4)\n", + "memory usage: 1.4+ KB\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "wine['Preference'].value_counts()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "G4IfBAQtj2Oc", + "outputId": "51fbc5a9-2080-4637-8ee6-164b59146f40" + }, + "execution_count": 167, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "Red 6\n", + "White 6\n", + "Red-Fruity 5\n", + "Red-Earthy 5\n", + "White-Crisp 3\n", + "White-Creamy 3\n", + "Red-Spicy 3\n", + "Red-Rich 3\n", + "White-Floral 3\n", + "White-Citrus 3\n", + "Name: Preference, dtype: int64" + ] + }, + "metadata": {}, + "execution_count": 167 + } + ] + }, + { + "cell_type": "code", + "source": [ + "wine['Red_Wine'].value_counts()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "vrGJsrb0j7Qk", + "outputId": "067da933-1552-49e0-aed2-347d1641e142" + }, + "execution_count": 168, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "Light-Bodied 5\n", + "Full-Bodied 5\n", + "Dry 4\n", + "Sweet 4\n", + "Name: Red_Wine, dtype: int64" + ] + }, + "metadata": {}, + "execution_count": 168 + } + ] + }, + { + "cell_type": "code", + "source": [ + "wine['White_Wine'].value_counts()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "bR5Ie3bakC1E", + "outputId": "3c9a81dc-ad0c-4a38-e4d7-4073bd222b37" + }, + "execution_count": 169, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "Dry 3\n", + "Sweet 3\n", + "Name: White_Wine, dtype: int64" + ] + }, + "metadata": {}, + "execution_count": 169 + } + ] + }, + { + "cell_type": "code", + "source": [ + "wine['Recommendation'].value_counts()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "iATAENA3kGN0", + "outputId": "92f4684d-27f4-4636-c766-caa0f13e575b" + }, + "execution_count": 170, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "Pinot Noir 8\n", + "Shiraz or Zinfandel 5\n", + "Sauvignon Blanc 5\n", + "Gewurztraminer 4\n", + "Cabernet Sauvignon 4\n", + "Chianti 3\n", + "Chardonnay 3\n", + "Riesling 3\n", + "Name: Recommendation, dtype: int64" + ] + }, + "metadata": {}, + "execution_count": 170 + } + ] + }, + { + "cell_type": "code", + "source": [ + "wine.isnull().sum()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "hXVqoyd0kKiE", + "outputId": "25a594a3-90fc-48e8-eb96-44d46d1f4dbc" + }, + "execution_count": 171, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "Preference 0\n", + "Red_Wine 22\n", + "White_Wine 34\n", + "Recommendation 5\n", + "dtype: int64" + ] + }, + "metadata": {}, + "execution_count": 171 + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "## Handing Null Values" + ], + "metadata": { + "id": "iaIBH_PQkU2l" + } + }, + { + "cell_type": "code", + "source": [ + "# Recommendation\n", + "'''\n", + "1.it is the class label and the dataset is also small\n", + "2. so i have planned to replace tha null values using the most frequent values(i.e Pinot Noir)\n", + "'''\n", + "wine['Recommendation'] = wine['Recommendation'].fillna(wine['Recommendation'].value_counts().sort_values(ascending=False).keys()[0])" + ], + "metadata": { + "id": "8fcr-w1EkTGc" + }, + "execution_count": 172, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "wine['Recommendation'].value_counts().sort_values(ascending=False).keys()[0]" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 36 + }, + "id": "ma_R478MlDk-", + "outputId": "0d5f3e9a-1eb0-4a09-99fb-8145b1147089" + }, + "execution_count": 173, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "'Pinot Noir'" + ], + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "string" + } + }, + "metadata": {}, + "execution_count": 173 + } + ] + }, + { + "cell_type": "code", + "source": [ + "wine['Recommendation'].isnull().sum()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "yX8GGdHrlFoF", + "outputId": "5579cf7d-2f7b-4489-d09f-f424afefd402" + }, + "execution_count": 174, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "0" + ] + }, + "metadata": {}, + "execution_count": 174 + } + ] + }, + { + "cell_type": "code", + "source": [ + "# White_Wine\n", + "'''\n", + "1. White_Wine has 2 classes [Dry , Sweet]\n", + "2. Replace with 'None' as a New label\n", + "'''\n", + "\n", + "wine['White_Wine'] = wine['White_Wine'].fillna('None')\n", + "wine['White_Wine'].isnull().sum()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "ZD3sh5V4lY-s", + "outputId": "621baedb-3daf-49a7-87e7-45bde6701212" + }, + "execution_count": 175, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "0" + ] + }, + "metadata": {}, + "execution_count": 175 + } + ] + }, + { + "cell_type": "code", + "source": [ + "# Red_Wine\n", + "'''\n", + "1. 22 null values\n", + "2. replace with 'None'\n", + "'''\n", + "\n", + "wine['Red_Wine'] = wine['Red_Wine'].fillna('None')\n", + "wine['Red_Wine'].isnull().sum()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "gshPG42-nTgE", + "outputId": "b9a2554f-2c33-4cda-9f2d-895f1a290f2a" + }, + "execution_count": 176, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "0" + ] + }, + "metadata": {}, + "execution_count": 176 + } + ] + }, + { + "cell_type": "code", + "source": [ + "wine.head()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 206 + }, + "id": "s7Ip6blqnlGU", + "outputId": "557ecc39-366c-4a2f-bfbc-db313ff9b10a" + }, + "execution_count": 177, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " Preference Red_Wine White_Wine Recommendation\n", + "0 Red Light-Bodied None Pinot Noir\n", + "1 Red Full-Bodied None Shiraz or Zinfandel\n", + "2 White None Dry Sauvignon Blanc\n", + "3 White None Sweet Gewurztraminer\n", + "4 Red-Fruity None None Pinot Noir" + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
PreferenceRed_WineWhite_WineRecommendation
0RedLight-BodiedNonePinot Noir
1RedFull-BodiedNoneShiraz or Zinfandel
2WhiteNoneDrySauvignon Blanc
3WhiteNoneSweetGewurztraminer
4Red-FruityNoneNonePinot Noir
\n", + "
\n", + " \n", + " \n", + " \n", + "\n", + " \n", + "
\n", + "
\n", + " " + ] + }, + "metadata": {}, + "execution_count": 177 + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "### One Hot Encoding" + ], + "metadata": { + "id": "Eehc1X-npf7-" + } + }, + { + "cell_type": "code", + "source": [ + "from sklearn.preprocessing import OneHotEncoder\n", + "wine = pd.get_dummies(wine, columns=['Preference', 'Red_Wine','White_Wine'])" + ], + "metadata": { + "id": "KsTVHcsvnnQb" + }, + "execution_count": 178, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "wine.head()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 305 + }, + "id": "4niIcd36pyUt", + "outputId": "410a9df9-294d-403f-b315-10350964c8db" + }, + "execution_count": 179, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " Recommendation Preference_Red Preference_Red-Earthy \\\n", + "0 Pinot Noir 1 0 \n", + "1 Shiraz or Zinfandel 1 0 \n", + "2 Sauvignon Blanc 0 0 \n", + "3 Gewurztraminer 0 0 \n", + "4 Pinot Noir 0 0 \n", + "\n", + " Preference_Red-Fruity Preference_Red-Rich Preference_Red-Spicy \\\n", + "0 0 0 0 \n", + "1 0 0 0 \n", + "2 0 0 0 \n", + "3 0 0 0 \n", + "4 1 0 0 \n", + "\n", + " Preference_White Preference_White-Citrus Preference_White-Creamy \\\n", + "0 0 0 0 \n", + "1 0 0 0 \n", + "2 1 0 0 \n", + "3 1 0 0 \n", + "4 0 0 0 \n", + "\n", + " Preference_White-Crisp Preference_White-Floral Red_Wine_Dry \\\n", + "0 0 0 0 \n", + "1 0 0 0 \n", + "2 0 0 0 \n", + "3 0 0 0 \n", + "4 0 0 0 \n", + "\n", + " Red_Wine_Full-Bodied Red_Wine_Light-Bodied Red_Wine_None Red_Wine_Sweet \\\n", + "0 0 1 0 0 \n", + "1 1 0 0 0 \n", + "2 0 0 1 0 \n", + "3 0 0 1 0 \n", + "4 0 0 1 0 \n", + "\n", + " White_Wine_Dry White_Wine_None White_Wine_Sweet \n", + "0 0 1 0 \n", + "1 0 1 0 \n", + "2 1 0 0 \n", + "3 0 0 1 \n", + "4 0 1 0 " + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
RecommendationPreference_RedPreference_Red-EarthyPreference_Red-FruityPreference_Red-RichPreference_Red-SpicyPreference_WhitePreference_White-CitrusPreference_White-CreamyPreference_White-CrispPreference_White-FloralRed_Wine_DryRed_Wine_Full-BodiedRed_Wine_Light-BodiedRed_Wine_NoneRed_Wine_SweetWhite_Wine_DryWhite_Wine_NoneWhite_Wine_Sweet
0Pinot Noir100000000000100010
1Shiraz or Zinfandel100000000001000010
2Sauvignon Blanc000001000000010100
3Gewurztraminer000001000000010001
4Pinot Noir001000000000010010
\n", + "
\n", + " \n", + " \n", + " \n", + "\n", + " \n", + "
\n", + "
\n", + " " + ] + }, + "metadata": {}, + "execution_count": 179 + } + ] + }, + { + "cell_type": "code", + "source": [ + "y = wine['Recommendation']\n", + "y" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "zIUiu-Wwp40s", + "outputId": "873fdc96-b9dc-49de-bfc9-28f4b58e5f6a" + }, + "execution_count": 180, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "0 Pinot Noir\n", + "1 Shiraz or Zinfandel\n", + "2 Sauvignon Blanc\n", + "3 Gewurztraminer\n", + "4 Pinot Noir\n", + "5 Chianti\n", + "6 Sauvignon Blanc\n", + "7 Chardonnay\n", + "8 Shiraz or Zinfandel\n", + "9 Cabernet Sauvignon\n", + "10 Gewurztraminer\n", + "11 Riesling\n", + "12 Pinot Noir\n", + "13 Chianti\n", + "14 Sauvignon Blanc\n", + "15 Chardonnay\n", + "16 Shiraz or Zinfandel\n", + "17 Cabernet Sauvignon\n", + "18 Gewurztraminer\n", + "19 Riesling\n", + "20 Pinot Noir\n", + "21 Shiraz or Zinfandel\n", + "22 Pinot Noir\n", + "23 Cabernet Sauvignon\n", + "24 Sauvignon Blanc\n", + "25 Pinot Noir\n", + "26 Pinot Noir\n", + "27 Chardonnay\n", + "28 Pinot Noir\n", + "29 Shiraz or Zinfandel\n", + "30 Pinot Noir\n", + "31 Cabernet Sauvignon\n", + "32 Pinot Noir\n", + "33 Gewurztraminer\n", + "34 Sauvignon Blanc\n", + "35 Riesling\n", + "36 Pinot Noir\n", + "37 Pinot Noir\n", + "38 Chianti\n", + "39 Pinot Noir\n", + "Name: Recommendation, dtype: object" + ] + }, + "metadata": {}, + "execution_count": 180 + } + ] + }, + { + "cell_type": "code", + "source": [ + "wine.columns" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "w9U_0ZcYsHW0", + "outputId": "9507b818-c4cb-444b-8994-01daadf6d7ec" + }, + "execution_count": 181, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "Index(['Recommendation', 'Preference_Red', 'Preference_Red-Earthy',\n", + " 'Preference_Red-Fruity', 'Preference_Red-Rich', 'Preference_Red-Spicy',\n", + " 'Preference_White', 'Preference_White-Citrus',\n", + " 'Preference_White-Creamy', 'Preference_White-Crisp',\n", + " 'Preference_White-Floral', 'Red_Wine_Dry', 'Red_Wine_Full-Bodied',\n", + " 'Red_Wine_Light-Bodied', 'Red_Wine_None', 'Red_Wine_Sweet',\n", + " 'White_Wine_Dry', 'White_Wine_None', 'White_Wine_Sweet'],\n", + " dtype='object')" + ] + }, + "metadata": {}, + "execution_count": 181 + } + ] + }, + { + "cell_type": "code", + "source": [ + "X = wine[['Preference_Red', 'Preference_Red-Earthy',\n", + " 'Preference_Red-Fruity', 'Preference_Red-Rich', 'Preference_Red-Spicy',\n", + " 'Preference_White', 'Preference_White-Citrus',\n", + " 'Preference_White-Creamy', 'Preference_White-Crisp',\n", + " 'Preference_White-Floral', 'Red_Wine_Dry', 'Red_Wine_Full-Bodied',\n", + " 'Red_Wine_Light-Bodied', 'Red_Wine_None', 'Red_Wine_Sweet',\n", + " 'White_Wine_Dry', 'White_Wine_None', 'White_Wine_Sweet']]" + ], + "metadata": { + "id": "E8uyeERprdpL" + }, + "execution_count": 182, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "X.shape" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "M7cCKVNusSVS", + "outputId": "0acb54a5-a1d0-453d-8b34-a3dadccdfa4f" + }, + "execution_count": 183, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(40, 18)" + ] + }, + "metadata": {}, + "execution_count": 183 + } + ] + }, + { + "cell_type": "code", + "source": [ + "y.shape" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "JX6KNEplsT2T", + "outputId": "8df2bde9-41c9-4092-b1e1-65ed81edacf6" + }, + "execution_count": 184, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(40,)" + ] + }, + "metadata": {}, + "execution_count": 184 + } + ] + }, + { + "cell_type": "code", + "source": [ + "from sklearn.preprocessing import LabelEncoder\n", + "le = LabelEncoder()\n", + "\n", + "y = le.fit_transform(y)\n", + "y" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "KwAXTRavsXJ3", + "outputId": "3f6e1f7b-6d7f-4d68-852d-eed24a13720a" + }, + "execution_count": 185, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "array([4, 7, 6, 3, 4, 2, 6, 1, 7, 0, 3, 5, 4, 2, 6, 1, 7, 0, 3, 5, 4, 7,\n", + " 4, 0, 6, 4, 4, 1, 4, 7, 4, 0, 4, 3, 6, 5, 4, 4, 2, 4])" + ] + }, + "metadata": {}, + "execution_count": 185 + } + ] + }, + { + "cell_type": "code", + "source": [ + "X_train, X_test, y_train, y_test = train_test_split(X,y,test_size=0.3, random_state=42,stratify=y)" + ], + "metadata": { + "id": "-SjHHFu5shkT" + }, + "execution_count": 186, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "print(X_train.shape)\n", + "print(y_train.shape)\n", + "print(X_test.shape)\n", + "print(y_test.shape)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "POOvu3hcspbj", + "outputId": "a82dad4b-4f07-487b-fc62-0137459b0aeb" + }, + "execution_count": 187, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "(28, 18)\n", + "(28,)\n", + "(12, 18)\n", + "(12,)\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "clf1 = tree.DecisionTreeClassifier()" + ], + "metadata": { + "id": "foK7ookRsxPj" + }, + "execution_count": 188, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "# Fit the classifier to the training data\n", + "clf1.fit(X_train, y_train)\n", + "\n", + "# Predict the class labels for the test data\n", + "y_pred = clf1.predict(X_test)\n", + "\n", + "# Calculate the accuracy of the classifier\n", + "accuracy = clf1.score(X_test, y_test)\n", + "\n", + "# Print the results\n", + "print(\"Predicted class labels:\", y_pred)\n", + "print(\"Accuracy:\", accuracy)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "fk3ZoFmrs9VL", + "outputId": "edbe4fe4-5746-4746-de6f-560e87d3a60b" + }, + "execution_count": 189, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Predicted class labels: [1 4 4 4 0 0 0 1 0 4 4 4]\n", + "Accuracy: 0.4166666666666667\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "plt.figure(figsize=(30,20))\n", + "tree.plot_tree(clf1)\n", + "plt.show()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 880 + }, + "id": "6XSnZLsSs_Uz", + "outputId": "22680862-7af1-4dab-af22-d4cc6113fc6b" + }, + "execution_count": 191, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + } + ] + }, + { + "cell_type": "code", + "source": [ + "from sklearn.metrics import confusion_matrix\n", + "import seaborn as sns\n", + "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score\n", + "plt.figure(figsize=(12,8))\n", + "cm = confusion_matrix(y_test,y_pred)\n", + " \n", + "#Plot the confusion matrix.\n", + "sns.heatmap(cm,\n", + " annot=True,\n", + " fmt='g',\n", + " xticklabels=['0', '1' , '2' , '3' , '4' , '5' , '6' , '7'],\n", + " yticklabels=['0', '1' , '2' , '3' , '4' , '5' , '6' , '7'])\n", + "plt.ylabel('Prediction',fontsize=13)\n", + "plt.xlabel('Actual',fontsize=13)\n", + "plt.title('Confusion Matrix',fontsize=17)\n", + "plt.show()\n", + " \n", + " " + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 520 + }, + "id": "Wjy1kMqqtHbj", + "outputId": "59a66ab6-a859-4617-91e5-468e8ec69684" + }, + "execution_count": 192, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + } + ] + }, + { + "cell_type": "code", + "source": [ + "print(classification_report(y_test, y_pred))" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "ONCikocltz97", + "outputId": "762a6219-0f63-4fb0-9a3d-978787857cf2" + }, + "execution_count": 193, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + " precision recall f1-score support\n", + "\n", + " 0 0.25 1.00 0.40 1\n", + " 1 0.00 0.00 0.00 1\n", + " 2 0.00 0.00 0.00 1\n", + " 3 0.00 0.00 0.00 1\n", + " 4 0.67 1.00 0.80 4\n", + " 5 0.00 0.00 0.00 1\n", + " 6 0.00 0.00 0.00 1\n", + " 7 0.00 0.00 0.00 2\n", + "\n", + " accuracy 0.42 12\n", + " macro avg 0.11 0.25 0.15 12\n", + "weighted avg 0.24 0.42 0.30 12\n", + "\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "/usr/local/lib/python3.9/dist-packages/sklearn/metrics/_classification.py:1344: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n", + " _warn_prf(average, modifier, msg_start, len(result))\n", + "/usr/local/lib/python3.9/dist-packages/sklearn/metrics/_classification.py:1344: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n", + " _warn_prf(average, modifier, msg_start, len(result))\n", + "/usr/local/lib/python3.9/dist-packages/sklearn/metrics/_classification.py:1344: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n", + " _warn_prf(average, modifier, msg_start, len(result))\n" + ] + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "### K-Fold Cross Validation" + ], + "metadata": { + "id": "6mTCBRDdvsI5" + } + }, + { + "cell_type": "code", + "source": [ + "X = wine.drop('Recommendation',axis=1)\n", + "y = wine['Recommendation']" + ], + "metadata": { + "id": "ECTwhEZwwLkU" + }, + "execution_count": 195, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "from sklearn.model_selection import KFold\n", + "kf = KFold(n_splits=5, shuffle=True, random_state=42)\n", + "\n", + "# initialize an empty list to store the performance of each fold\n", + "scores = []\n", + "\n", + "# iterate over each fold\n", + "for train_index, test_index in kf.split(X):\n", + " # extract the training and testing sets\n", + " X_train, X_test = X.iloc[train_index], X.iloc[test_index]\n", + " y_train, y_test = y.iloc[train_index], y.iloc[test_index]\n", + " \n", + " # train the model\n", + " model = tree.DecisionTreeClassifier()\n", + " model.fit(X_train, y_train)\n", + "\n", + " # evaluate the performance on the test set and append the score to the list\n", + " score = model.score(X_test, y_test)\n", + " scores.append(score)" + ], + "metadata": { + "id": "qjtOJfy-t_kj" + }, + "execution_count": 196, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "scores" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "4TLrmi_rwX5j", + "outputId": "01b8496d-31d7-4aa9-8518-a2c24a54f0dd" + }, + "execution_count": 197, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "[0.375, 0.0, 0.25, 0.25, 0.25]" + ] + }, + "metadata": {}, + "execution_count": 197 + } + ] + }, + { + "cell_type": "code", + "source": [ + "\n", + "# calculate the average score across all folds\n", + "mean_score = sum(scores) / len(scores)\n", + "\n", + "print('Average score across all folds:', mean_score)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "RaaNfImIwDej", + "outputId": "2595e52e-80d3-4d6a-89a7-7d6c955a914e" + }, + "execution_count": 198, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Average score across all folds: 0.225\n" + ] + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "## Use Stratified K-Fold" + ], + "metadata": { + "id": "05mPzYQozTA1" + } + }, + { + "cell_type": "code", + "source": [ + "from sklearn.model_selection import StratifiedKFold\n", + "\n", + "# X is your feature matrix, y is your target variable\n", + "skf = StratifiedKFold(n_splits=5)\n", + "skf_score = []\n", + "\n", + "for train_index, test_index in skf.split(X, y):\n", + " X_train, X_test = X.iloc[train_index], X.iloc[test_index]\n", + " y_train, y_test = y.iloc[train_index], y.iloc[test_index]\n", + "\n", + " # Train and evaluate your model using the training and test sets\n", + " model = tree.DecisionTreeClassifier()\n", + " model.fit(X_train, y_train)\n", + " score = model.score(X_test, y_test)\n", + " skf_score.append(score)\n", + " print(\"Test score: {}\".format(score))" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "Gvzhfx_owoUT", + "outputId": "dfca45db-9c49-488b-cc71-45e89c52f2ce" + }, + "execution_count": 201, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Test score: 0.125\n", + "Test score: 0.375\n", + "Test score: 0.25\n", + "Test score: 0.25\n", + "Test score: 0.5\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "/usr/local/lib/python3.9/dist-packages/sklearn/model_selection/_split.py:700: UserWarning: The least populated class in y has only 3 members, which is less than n_splits=5.\n", + " warnings.warn(\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "skf_score" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "3lYM1Kguzj6j", + "outputId": "95c02fca-e8d8-4d60-eb26-9976926186fe" + }, + "execution_count": 202, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "[0.125, 0.375, 0.25, 0.25, 0.5]" + ] + }, + "metadata": {}, + "execution_count": 202 + } + ] + }, + { + "cell_type": "code", + "source": [ + "mean_score_skf = sum(skf_score) / len(skf_score)\n", + "\n", + "print('Average score across all folds(skf):', mean_score_skf)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "5v9a2zth0GEC", + "outputId": "24263aaa-8693-4cc9-eaf1-190bdd135627" + }, + "execution_count": 203, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Average score across all folds(skf): 0.3\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "new_data = [['Red','Light-Bodied','None']]" + ], + "metadata": { + "id": "xpP2OCHi1AF6" + }, + "execution_count": 216, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "from sklearn.preprocessing import OneHotEncoder\n", + "ohe = OneHotEncoder()\n", + "new_data_transformed = ohe.fit_transform(new_data).reshape(1,-1)" + ], + "metadata": { + "id": "GOhB6sZf0OsR" + }, + "execution_count": 219, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "new_data_transformed.shape" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "5w27VUrH039B", + "outputId": "c83493ba-ab65-4ed7-8467-dcbef5776ef3" + }, + "execution_count": 220, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "(1, 3)" + ] + }, + "metadata": {}, + "execution_count": 220 + } + ] + }, + { + "cell_type": "code", + "source": [ + "new_data_transformed[0][0]" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "gmkyYi7r3aKa", + "outputId": "0e1def97-7c5d-4808-88c5-2c8a930ce711" + }, + "execution_count": 226, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "<1x3 sparse matrix of type ''\n", + "\twith 3 stored elements in Compressed Sparse Row format>" + ] + }, + "metadata": {}, + "execution_count": 226 + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "### Using Pipeline" + ], + "metadata": { + "id": "50I5UMNa5MPJ" + } + }, + { + "cell_type": "code", + "source": [ + "wine = pd.read_excel('/content/wine_data.xlsx')" + ], + "metadata": { + "id": "EcbyQ1vW5N4J" + }, + "execution_count": 240, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "wine['Recommendation'] = wine['Recommendation'].fillna(wine['Recommendation'].value_counts().sort_values(ascending=False).keys()[0])\n", + "wine['White_Wine'] = wine['White_Wine'].fillna('None')\n", + "wine['Red_Wine'] = wine['Red_Wine'].fillna('None')" + ], + "metadata": { + "id": "sEfcQb5a5epv" + }, + "execution_count": 241, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "X = wine.drop('Recommendation',axis=1)\n", + "y = wine['Recommendation']" + ], + "metadata": { + "id": "Lfk7Ha6p5-jK" + }, + "execution_count": 242, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "le = LabelEncoder()\n", + "y = le.fit_transform(y)" + ], + "metadata": { + "id": "q4S2u4Js8Muq" + }, + "execution_count": 244, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "y" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "lSNmXh-g8UYx", + "outputId": "05a0768e-4f5e-4aac-a88a-6d8c4fe013d8" + }, + "execution_count": 245, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "array([4, 7, 6, 3, 4, 2, 6, 1, 7, 0, 3, 5, 4, 2, 6, 1, 7, 0, 3, 5, 4, 7,\n", + " 4, 0, 6, 4, 4, 1, 4, 7, 4, 0, 4, 3, 6, 5, 4, 4, 2, 4])" + ] + }, + "metadata": {}, + "execution_count": 245 + } + ] + }, + { + "cell_type": "code", + "source": [ + "X_train, X_test, y_train, y_test = train_test_split(\n", + " X, y, test_size=0.2, random_state=42,stratify=y)" + ], + "metadata": { + "id": "3fx7c9hY5wha" + }, + "execution_count": 243, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "from sklearn.pipeline import Pipeline\n", + "pipeline = Pipeline([\n", + " ('onehot', OneHotEncoder(handle_unknown='ignore')),\n", + " ('classifier', tree.DecisionTreeClassifier())\n", + "])" + ], + "metadata": { + "id": "0G5B1atJ6uhy" + }, + "execution_count": 247, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "pipeline.fit(X_train, y_train)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 126 + }, + "id": "wBakGDQX638K", + "outputId": "0ba601ab-befd-42ce-fbfa-b96df3c886fb" + }, + "execution_count": 248, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "Pipeline(steps=[('onehot', OneHotEncoder(handle_unknown='ignore')),\n", + " ('classifier', DecisionTreeClassifier())])" + ], + "text/html": [ + "
Pipeline(steps=[('onehot', OneHotEncoder(handle_unknown='ignore')),\n",
+              "                ('classifier', DecisionTreeClassifier())])
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" + ] + }, + "metadata": {}, + "execution_count": 248 + } + ] + }, + { + "cell_type": "code", + "source": [ + "y_pred = pipeline.predict(X_test)\n", + "\n", + "# Evaluate the accuracy of your model\n", + "accuracy = accuracy_score(y_test, y_pred)\n", + "print(f\"Accuracy: {accuracy}\")" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "swEZCh2_65SR", + "outputId": "bac8fba1-51f6-4f7b-980a-41e8ab182e4a" + }, + "execution_count": 249, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Accuracy: 0.375\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "from sklearn.model_selection import cross_val_score\n", + "cross_val_score(pipeline,X_train,y_train,cv=5,scoring='accuracy').mean()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "K8DDm4r1ADEp", + "outputId": "101bcb1b-170b-495f-e568-ec12b645e567" + }, + "execution_count": 251, + "outputs": [ + { + "output_type": "stream", + "name": "stderr", + "text": [ + "/usr/local/lib/python3.9/dist-packages/sklearn/model_selection/_split.py:700: UserWarning: The least populated class in y has only 2 members, which is less than n_splits=5.\n", + " warnings.warn(\n" + ] + }, + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "0.2857142857142857" + ] + }, + "metadata": {}, + "execution_count": 251 + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "### GridSearchCV" + ], + "metadata": { + "id": "4FPMpMDHAm8g" + } + }, + { + "cell_type": "code", + "source": [ + "params = {'classifier__criterion':['gini','entropy'],'classifier__max_depth':[4,5,6,7,8,9,10,11,12,15,20,30,40,50,70,90,120,150]}" + ], + "metadata": { + "id": "xd41Bw22BN7r" + }, + "execution_count": 255, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "from sklearn.model_selection import GridSearchCV\n", + "grid = GridSearchCV(pipeline,params,cv=5,scoring='accuracy')\n", + "grid.fit(X_train,y_train)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 208 + }, + "id": "XppiX9naAmaR", + "outputId": "729ad00f-86bf-4d17-9d30-18694fcc6d98" + }, + "execution_count": 256, + "outputs": [ + { + "output_type": "stream", + "name": "stderr", + "text": [ + "/usr/local/lib/python3.9/dist-packages/sklearn/model_selection/_split.py:700: UserWarning: The least populated class in y has only 2 members, which is less than n_splits=5.\n", + " warnings.warn(\n" + ] + }, + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "GridSearchCV(cv=5,\n", + " estimator=Pipeline(steps=[('onehot',\n", + " OneHotEncoder(handle_unknown='ignore')),\n", + " ('classifier',\n", + " DecisionTreeClassifier())]),\n", + " param_grid={'classifier__criterion': ['gini', 'entropy'],\n", + " 'classifier__max_depth': [4, 5, 6, 7, 8, 9, 10, 11, 12,\n", + " 15, 20, 30, 40, 50, 70, 90,\n", + " 120, 150]},\n", + " scoring='accuracy')" + ], + "text/html": [ + "
GridSearchCV(cv=5,\n",
+              "             estimator=Pipeline(steps=[('onehot',\n",
+              "                                        OneHotEncoder(handle_unknown='ignore')),\n",
+              "                                       ('classifier',\n",
+              "                                        DecisionTreeClassifier())]),\n",
+              "             param_grid={'classifier__criterion': ['gini', 'entropy'],\n",
+              "                         'classifier__max_depth': [4, 5, 6, 7, 8, 9, 10, 11, 12,\n",
+              "                                                   15, 20, 30, 40, 50, 70, 90,\n",
+              "                                                   120, 150]},\n",
+              "             scoring='accuracy')
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
" + ] + }, + "metadata": {}, + "execution_count": 256 + } + ] + }, + { + "cell_type": "code", + "source": [ + "grid.best_score_" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "Rmghi5mxBnma", + "outputId": "c2287da2-f266-4606-8197-51d389eea71e" + }, + "execution_count": 258, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "0.35238095238095235" + ] + }, + "metadata": {}, + "execution_count": 258 + } + ] + }, + { + "cell_type": "code", + "source": [ + "grid.best_params_" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "6oesag6TBs-A", + "outputId": "6dc525d8-8a24-4678-81fd-addab9e346c7" + }, + "execution_count": 259, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "{'classifier__criterion': 'gini', 'classifier__max_depth': 4}" + ] + }, + "metadata": {}, + "execution_count": 259 + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "## Exporting the PipeLine" + ], + "metadata": { + "id": "ncTd9LuTBxzD" + } + }, + { + "cell_type": "code", + "source": [ + "import pickle\n", + "pickle.dump(pipeline,open('/content/sample_data/pipe.pkl','wb'))" + ], + "metadata": { + "id": "6lNtvIt7B5fa" + }, + "execution_count": 260, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "pipe = pickle.load(open('/content/sample_data/pipe.pkl','rb'))" + ], + "metadata": { + "id": "qREsT8mGCIvL" + }, + "execution_count": 261, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "new_data = np.array(['Red','Light-Bodied','None'],dtype='object').reshape(1,3)" + ], + "metadata": { + "id": "mrZsYNQF68zR" + }, + "execution_count": 265, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "new_data" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "-m-6qe2GCk5o", + "outputId": "85d423c3-5a1e-4072-8f99-6947d8a2dd18" + }, + "execution_count": 266, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "array([['Red', 'Light-Bodied', 'None']], dtype=object)" + ] + }, + "metadata": {}, + "execution_count": 266 + } + ] + }, + { + "cell_type": "code", + "source": [ + "pipe.predict(new_data)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "0y6C3YUjC41h", + "outputId": "01b5ba9b-ad77-4e84-c2c2-1eddced5e38b" + }, + "execution_count": 267, + "outputs": [ + { + "output_type": "stream", + "name": "stderr", + "text": [ + "/usr/local/lib/python3.9/dist-packages/sklearn/base.py:439: UserWarning: X does not have valid feature names, but OneHotEncoder was fitted with feature names\n", + " warnings.warn(\n" + ] + }, + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "array(['Pinot Noir'], dtype=object)" + ] + }, + "metadata": {}, + "execution_count": 267 + } + ] + }, + { + "cell_type": "code", + "source": [], + "metadata": { + "id": "-yrFSUIi7Xgx" + }, + "execution_count": null, + "outputs": [] + } + ] +} \ No newline at end of file diff --git a/pipe.pkl b/pipe.pkl new file mode 100644 index 0000000000000000000000000000000000000000..a614ec7ad426e97f242a544678abc8e3b76d4c45 GIT binary patch literal 6191 zcmd5AO>f*p)M>KGW}BvI+CqTRE@_2=pb;QIl>pglpdbuOX%R?8$a1`sjD^=W<8ez6 zNN{LMq~QXCI3vD_gt%4W2;#s`fCEB{pp=hG4?5X<=K|uga6a~968IsJqJ9z|-;}q!Dqog!qLy$HC-PZ%9F0OkBQC|Eo+cwP zi#(Dfez-jEkkE_Tge6k!dORc#NBqH@AbZvtp3BK{#C!rmBW}tgDMpjnWeIq1w80ut z5^b=MQ=KT~amwXuqSodsF-XrjjYD(iIBTIxqR;que zNZWZjYh~kU>T{AK1cb%T^<*8@PTHii$~`@T&KrQ-ni(AiJJ1jKfAm3@aSC~4$O)GE z91!K_=qcZ$QjDv|)3Fal6}8MKIS}>*%%jt6jtZi&il{w3lK0+U(EhR+6jH717 z@!Kyr?h+@=S!Pih!(hX-jQ3c8OgVic0rQz{huyGi_AmjaUJmChY=R%bFxKvvzjOXh znJ%tE8NI-TZCUaOlVU&a5od(VI~hwY#x)NR&sVxn0)1g!p|?ik9)sqx%=*Z(ANtUe zBo2HIM40(SlE9XZs|kmd;uK=uqvk(xJ8=L5n~6$$hnVW8!ZBEq_K=uS#B)Ry6*k88 zIJj{#=0o}AFbpxFY~xzsHYph=gbR=WWrD#Nu+M<+S62N_oI>O)5S9ni088Iz zGD}}O28D6Yiyb$Jsk<&j57^V95$dCmz%mYrx-Od|tAkU=K8xuEcagA=@aAbZJ*%z* zG^VjYUDl4mP%Y}%0`=XHLn$X8BIi=}B4;kdtP0vV>4y<-wjv*p`YAYGa4wgn*xj?$ zJRZ2AC&k{=)MxIC&4@M6z%_vqkbJ0B)a!vBgrn!Vfy>lEm=inv!mi&}+I0iQbD>_> z#h0saE8w2>H2~@`Eoy3r94V%YL#3!N5^9pbnLS-C(hwABb*;|{1+gfm>0rR%9md3{P6v4c5 z3OANUN~VsJ0^JC+nZn_f;@W&R;nI~jJPY{UazW{Z;^|c7w2f6s-=m+{b%5^!oC5fP z(p>6m#fF+QAo(G^WH$j`1kE{s9|3;*s!Z);aKop-{@U@2V}>!LTXs;xowz?4?3x*9 z{$e%@bj#YI@f$Q8E%3YM$I%;qeD(QHCh~Lkz1FanyGFxWN$eN@A^!@_(}epi9t zHR;E@U;q7`@65>p{WeRz{`xHhqqS*k^6t!ui0wi~O#cweP?c6YH1j%~%uK0M-%bE+==*##isg z_cyMX<+}9OQax{6&!0@3kKEqW!1)#9xM~jOZ+&laf7{o?xtrGY!9HSL@_qQm#J$Ps z^?~y%_Qx1luUr@0tNs7;IIq?5U9Tb5H%Ik7Gk&Pd3*MXSv>xsu>ixSubj|;&58Iv3 z13Ir8HT=KM*&!X@tcJz8@0y=xhyL5jooH(Ln=~xWFZTWGO5X>qXEBb!_pcbo_V0v* zH?|l(>jo)&b=}}=4ZPIr7bLy3cKoy6sjpjkF%TLh{AP_KeM;j9ujX%mW2^e8hwvZu z^DujUP!A!-SBAa#wLZ$1VOdX$8%K1$h?n&cM}E(aCd^trPWFKE2#ee*w$G~4`Y9DV zC3)-<>Y#iXBEKAeSs&$4AMs70wJ9fKRP2=H%eXd&dI&YyD92yn|AT&6zKo-vtAJ~7{ft3^ft$x|;R`Egom*vWM&#%>cDm-i| zUyuUCDid;iQ2jyWHuho|gX#||x0#;`-O4H3wyX6kcDG2V_}`{od|y`V4HysV4=7#{ uL;dagM}ynfSKoHhMS6xlqkfSjVzfg#5&W#$5@Ya__x8Y$GIYa;fFL1V(%p^HDcw0V2sm^~NrQlZv^0Wrmvl)864KrM8$IX! zJco0>zukPgFAVL`zh7*@o#-ONr_w z+K0#zrvrf->30&)oMQFJ%II*;Od|BXJ*2T}Wcg1sMod3hh7yX5T1Yre3)PF|XJoVasB;d|S3hAod_sta4VoLvnLJB<9ES^?je%9<Udn2*L1m2(r3{!F3j<0mtR zSsnSFigdo228^Xj;t!PaZS=YBo1N#6-0Yz^UE(MX$HC)!T;o;}lz8drf=vI!DM8Y) z_%_^#X|j~82d!)48~UP8`B`$k3y&qo4#dl`hL|)+P;qm?K?DNHueG}rw7wag zm%>emzSt=VDyw~#xfwT<>N}N?w}CDc!Xv*unvCBAF*5&J>fK{QadJVT_QHb4vdk#M ziI>vDz{IBINIbO}_r{x1F|Fr0B?rzq+mKW@b>@k$b`{4#f4WD#k(trd2V8Uj`-Ga)xu!4hH5 z_G^;YqSLA3zeUki?5LstlC(xGT&&qO{sKKLf*->y6^d5T*RmsU8qw%C8;m?V%)O)d zl6RKgynHeC;{>9x4*CZBUAe8{^%I1vBO-55Z#BXz6W(Ac^|m{zr*|Q%?^?=kOgTWK zh1EF1`+}iA9@g<#IMkUDH(KXdM^{`cIHebWh>2sYVp$~Ouv`rL+>WaN`k{qjH3neO z@^VyUajVK81P-_7hUJASp7;nWGoJL0@J&%gVR<(lGB)Ydu~H{OU?mU--Yt{bQCQtc zQPoP75^aQA^PB9x1CQ<;Qv&u|ozDJM0kA-wcdn2V)2XjAd8iA3!J(dFGGYN4a2}ee z_Ei@Z*j2esBkn0eMzI?`zv94Pj-@&0L2HD)7`#@_w{viR zNxB`zdkb)$o)t@pjH!JUa_Z<6BC1BeT`dL^YIX1x`|;*drt|&#)>HraWwf!et1C0<%8z zQoOR^dv${}U-a(zgg|}-oWS->d+EHJg4gJQJmDOdmt8k&x7~_-15=L0$_Czs?4#>9 zo*Ll-7QPfIrm_L`*i_%z-mImRN{K`BNP_*Cr7KKrs}0cqJ!8BUj88m*k)R$%2oeAa z9E=QqB#pmP;y(!l4yNy5l=|;Jik~aUb-?ru+HEk4TdFHQ_E%?S%5AkROqAX#y4fcb z>^{c}r1bTAnv-(OK>Hw%y)WIaXPg*IK-`nor}<$x2%fk$dwhuYUyi;2QH*v<$e>b) zxH#LJTSYr)7%33kI>Fc;h73IJALq2x#KG_LD8-NKMuRx423cI#tRTY=`(OdIziP$r zGT_B_3AC!VmOO>dZ1t{AZ1xoC9n|B|$PqEB3_rRDU}Sxq;^}#-#3{;Q!?l@$$DnoV z`%BO&xJZtykW;`r$eBF2N9vI;T`XiZJNdKP-gjrwt3JBs*+Yi@<2|#L-tLJ7*h>F1 zev;$1u+qZz0o23*01@npKjSCF!qn6m!u)Gs{TWEp;`HoOS@44PXm*4&n=p|;9GTJd z3aTY;WA!CdQLL>Gx&~aEICh%%UM5sMAEzrsn4T6K^3oqPboEtboIx|r-IWC+Qb=Ym z*T>H1bqF;HDdrJrb=QYg*W*$PqK5CSy%YL^F=J!ru^UR5xeR6ZCp3Q8|APPS* z9-KOb#Nu`_=UCr8Gt3SkDo+%HGf+9`ejYZ8_S7eE1hcII&E{M;Z@RazqnVIH?}=7Z z%CG~8!nIAzkN#e<*x_x8K?@(ko7N*O>Z~^ZrzYZhOA@NZu_$pD1J8z)=rB&dfO%?s z2uc;qSyNv~sp zn~6->ahs`n`^$Za9))FLZd9CoUJf~xboDl;_2*ca>wc^Ch)Db~vF!^;3SCjgH5E<@ zp>*g=7M_Vl9d47BCu~e!vxzME5t#l5EW6!fexuwuD%W??wEk}b4#3F^(wK~Hv>Q)^AUD@O{PI_4%T>+SV-=I|Rx~9N<6)Dd**n&*a$zGeMkh>k zQBNdF0{P;QL#Sd_q;otiDYengo?v6m;$%{~yyvplK4A*wPl+5!VGNlx!lih_Cq*_P zcQl+*5ckSc}Ns$xL433WXQqg{s-3P12uya0w^) zYA478Rola1AR)kGtifq-Yin&!x3U`n>4kZ#>2Pl(EiQopMUM$GOA9^;Vpc(elx+6q z7fOI{orU)^Wd=&pf?^TJhr=TKkFaPh;+?-&5lIgrVp?Zg%D@gq#07BptO<0i^Fru&DBt$G2hz50&A)VPkvi;4rz zVSSmdS51apvSb8d$%C52}RkI$wM5_My zB`-X8%M5i{C-KYWDvl~Nd_|-nNym0zJ4J+=F^<0&?@0DjbvPBjEXMH)MQ+jtI9;SV zr$b81kUvL0C4S`g7|!Y}94MSHK-Sx?!a6o~i(r)q;_uU3%CdphjetaHlzSy3$fCoN zGk(x0@5~&iIPnk*fboxH*|e*z7it$C2>sHCzfj8&+4=ysa~1cc-bU7x5RcAt!_2sb zzO{*y_!6`e(ro{9h0ET3AR@AIZn|=ky1_&;=Vn*Fl`}EJaAzTdmLF8+_JK03;~DSs zq%XsUNnN&QOy^ZEOP{wY7saITq~UM1OT3_J&I6WQ?sNJwNg#~D#OFVwtBYW=WgJYI zMv(lhwET*$&K9P&rp&+Ie+AZUEuHXLNjx9sVRvc~K zl=Mb%NV0#6M1390f0C;#EoboH)s`3?I>+#E;%!o30M`w+0DEjn@(Fipe_j=}wPwzU z+n_gOaO4@GUV}4p-XOLjkD<(LiGKbk48<-R(O9k7Mow^iN@+Wy?30iRgxY!PAG^8p zZ-nYYT*U*7xBXtVxhc;^1|;pp_j`lBgc!Vpq_tGi$r)2Ea;3keq2onrW@VhBf13$@ z9Q#QTYUJScikI4W8O=C<0P^qmV^mn*{3lU&cZUDS{!SS@r>j%;+CN3?23R0!9V zzvC=mOSuD)-{Ld=$y$ONbHa$Cop>jK&KJAPTc^)+FP$`ByvnN_sedta)RXBtLBwji;6@tguGl2_cwam4V1_Pcs2ljQZ5HX2B`KA*1ZH6qFk6e7#DH)D;!Vr5W9XkskAqGa=%N(h;8 z(6b5JcYQ!_4Aj@3z9EVwH>B}?8nDYQ`I(^_X|ZryE_~l1W?(!01KLA3+^oks z3ly>u2xo+lXbmF>vE+27hkv$sCb43f!wS1=NV%!A8w%@Ey;2w5vJHJ2U&jWe5qfWQ zM1&}PL}@h9VMP^ofGTPE*XA8(;_Ary;5YJA4Fn+N!n!W960dt`!pCF#`;(IQ zmSYvM)#-sV_R^m6vFs-|@fBBh5f!tXmtc6$GPDDNlq10YA6YkpX9bWXiWDS#l>(tgOnOiPFJEO}w7G|LPwrsUCXsDk8%2ctX zA0KpTj7gPpNVdlX%`r&yb;%^3M}(}Ci$XLPH72;6yC75Sy930*F3nczq($R2IN$tGK1D_w}m!>`CJ`7*L}cvJ^V-!j-D5uT*}p z&dgJXr9dC9EKE~PIA$YZrQXHe$yuwQQ8Eu5N4qYWs){XQ7-4Oq;89wdd8v+sEV?Hf zUQumU=7>ep8(leUo?LW}k+;w%gO0Q4n}TVPks}Q|iiQiAi3eb-;C@8rY<6yb(r^y1 z?+cPzKj;VrI^HuZ$~v=ghJ>~hh{?>kuI^7w2;=5bgy~cxLH9pWPr++l)W-Pdq{?)S z(UGAE>}ZEaIJ=@fcxM@SQo!cxRaRw!CVeKt!Q(3F9P>?F*~sTTJ&>CM9OYZ6#X*BG z$UY^iKE6;8N|&g2Bdju^JQ=mlFM1J?9qxgoL0O;oes_{}c=wQ1O!%1?cSCqx zB{XDGZamY;4bmubX^{Nz^^6zu&=eHGANPiz0878q`po`U;=Esm^vLqF$$_cC=de76 z{zndjID6QbLVj)wa@1_CXIZdr1*d(nE_eG}ml;rN`bWo=*Sj(wZWkH#hvw@?P?-;Q z-CilW>VtV}MtnjQ*Tx`2^N{DGsfrJ(jJupl*uZ*$M67C?VMPi-&4T!?pvPPJh55r_ zrtU)wO}z`vawW59r=Ls*DEySU+I7}a5GiWr*kpFz+9g4mtGQx%r&G|&BMqx6Zlm$2 zFxu&2;-B)9GniZmk;kU2)u75Ul5U=-pzWRt5J*<&&e_LvIWziM5h_@O)>~K^jVBw& ze!HT4vywZaKDGP6A+*$5zar6j%|y|{LlS;vib4Vvbvb;7??^TWmDQ%Z(VJVzx+F~^ z*X}Kl^>w~|>yp;SUg7?`21rQXi!0zxK0qlzCL=-H!XGS+oW3OxCd(y0_q=r~=&CG5 zmg&@=?=(t(M=Wh+*lUj}qua3DAq36Q2;U2SvFAnf5bOH~)0zG^RrchKP#Myfm!Eva z?LqTWj%>WbmjNF2lX=iy!L#M{z89@l$l~Y=HaDBjz-H?maQ?dw_#a41^7nH=HV{Q`?kNZFu(CK0oyinV)??>9!F0Sz|^@Ic!tLy~SKRT53VTC}`f zVv!mv=JNjO)HoK$ys$N+7dega!TGShnLpt{EO*@mr)DMh-#j~n`rn`+c2ht%#z0kO1) zZDbDzOGNC9r5X&TuRfn_kEi6cJUs9DRQ54xyW+x8`9sc==6*}VbSqzVwE`C+@`qhy zJaKyQfRsn32D$oJ^($DM=bP?=LIvYwdWe`yp-lW@zoWJ9QveIq<}sIK8&*L)3<3+8=}m zObf&})3+(7isowfKkopb1FY2NNkK8VGjU`bGM;&Hg z@S$>z*IQOrzk!#SYmSsQmG0bVCbSt8aJa(MxNx)h_hK%eEt^~`Y)wRA>+|@JwfR}O z`m-4EM=9%fN#a)_E8tCQSQ`tT#2N5b%-t()G7u#8;u+;9wKbs8z=x zt%;9T9;iZfL)7lm@wGe(wIEvX$5)w950>7GaMI!8LMz)t1LP`Z#p*qVl~)@6yaLa) zZnvXeJ>s`}-#jRNI20hm6a7*!TbidwV&~A$NMy_HG{gE8211_gK>yJx>c63LXi=)F z4R-eo_y7RSA2Z+pgQc^n(+g8)=b!jwzIV>bWMpZ#t@=fFPMJu6y?kdq4+C$`StaMG z<2op=U|uab$5nPTfQtQ-Bdhasl!)GFaq(Q}Ifcb+;mK>PVcbvt;93XK+ld6A=Or^hsNgfpo);|;pf z^uWutY5K&C+CCBuG>WfIr5maVSNL+qpAJn^4xfQj9~#`n^{#vGGA)BYR!?kR?X1C1GF~Pcxx^0@b!e8GA5~#b{XtTJWrf9Yx*{NSq zJq{PRTG{;Com)A1fH_%wKt}JT`=IFXZMnt;Z^?&4iTzvDs{tn0X`rh4M#VzD-D36- zZ9Cc73lPdJ;f4cq2KMZeZb8YFvx@*;C6|gcPU37oTJi@W@c%LlWVbNF7Wtq1MoaKG zo&*dATClu}4J+iD*c&T5**id(jqROG|Lo8IuYL=gI`7zit9BMFiL=04|B-JAHgOy% zvssuGR+-TO3A+ItGV+XC(pKr$uU4gW6XTgU&jyqE{3}*v)LRLObSCMg9L1FZVN=mB8S)TTS+*@h4?G$AOgnl+0+Q@K5HOz-sDSNRIYt0Q@SZ0UfvIMT?vmw*hoEs z50{#_VtMX3kc)}$ULv=kQa~75SzE#Fj4b1ivC*YlrE+dL)R3{jVu(2LmE2KA*UK(1 zb8afLUOZ9Ms;@ldLwEdO)9U5u{LY4lBsDKIr7HaUQTMs+O~Zydy4=LKFd#i!E8Vje zUE8p$FLrU?WwHeHJBxL$`1a(mFxOfkdS_`&u9!gVbMmS;uT=EMx&QOV)mnK*Lat84pVgh#Gbd5OkfBM zd)Tl9iS<|P89F%p4|*`n{duItirD`wi|zqWAA#mcdA=%xUKeLe^BtEN!o|9pE94nx zRZxW#$81l%sTmp9a+`@<@p8Wumzf9i85~ECbSj6gX5%3ukXjRBymg=RZM>v>K!t0y zZ0skS2I!2uJr}Ws$Vd0W5rDDFa2Qpz6lh_^sQQan-(6n z;rh~`;(Z*`+q7CM^T=RyP^de|J)=r=VlMvrz+QMC_qJO z&$x8d4o!bstF9TEoMPXu>j*|bR;*1@kTK8Mp!Jy+vlNgpv5Vp*yy15o$Gslltwqg~ zF`Z`i?O0SAg29Prgc%2{JIm=5?s{vh$~)R|Sw9H5 zKVI=GV>a||2S+Pdx4~^{40##4Uw_yL-da7$zE(Z_JZL7{bYpsh{2%nf!85|7+`m8F z^q5+hOX%Uk>kyvVRq6_nrPd x$^Ui<0AxP`0REBk@0