{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Regression with XGBoost\n", "> After a brief review of supervised regression, you'll apply XGBoost to the regression task of predicting house prices in Ames, Iowa. You'll learn about the two kinds of base learners that XGboost can use as its weak learners, and review how to evaluate the quality of your regression models. This is the Summary of lecture \"Extreme Gradient Boosting with XGBoost\", via datacamp.\n", "\n", "- toc: true \n", "- badges: true\n", "- comments: true\n", "- author: Chanseok Kang\n", "- categories: [Python, Datacamp, Machine_Learning]\n", "- image: images/xgb_feature_importance.png" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", "import xgboost as xgb\n", "\n", "plt.rcParams['figure.figsize'] = (7, 7)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Regression review\n", "- Common regression metrics\n", " - Root Mean Squared Error (RMSE)\n", " - Mean Absolute Erro (MAE)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Objective (loss) functions and base learners\n", "- Objective functions and Why we use them\n", " - Quantifies how far off a prediction is from the actual result\n", " - Measures the difference between estimated and true values for some collection of data\n", " - Goal: Find the model that yields the minimum value of the loss function\n", "- Common loss functions and XGBoost\n", " - Loss function names in xgboost:\n", " - reg:linear - use for regression problems\n", " - reg:logistic - use for classification problems when you want just decision, not probability\n", " - binary:logistic - use when you want probability rather than just decision\n", "- Base learners and why we need them\n", " - XGBoost involves creating a meta-model that is composed of many individual models that combine to give a final prediction\n", " - Individual models = base learners\n", " - Want base learners that when combined create final prediction that is **non-linear**\n", " - Each base learner should be good at distinguishing or predicting different parts of the dataset\n", " - Two kinds of base learners: tree and linear" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Decision trees as base learners\n", "It's now time to build an XGBoost model to predict house prices - not in Boston, Massachusetts, as you saw in the video, but in Ames, Iowa! This dataset of housing prices has been pre-loaded into a DataFrame called df. If you explore it in the Shell, you'll see that there are a variety of features about the house and its location in the city.\n", "\n", "In this exercise, your goal is to use trees as base learners. By default, XGBoost uses trees as base learners, so you don't have to specify that you want to use trees here with `booster=\"gbtree\"`.\n", "\n", "> Note: `reg:linear` is replaced with `reg:squarederror`" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "df = pd.read_csv('./dataset/ames_housing_trimmed_processed.csv')\n", "X, y = df.iloc[:, :-1], df.iloc[:, -1]" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "RMSE: 28106.463641\n" ] } ], "source": [ "from sklearn.model_selection import train_test_split\n", "from sklearn.metrics import mean_squared_error\n", "\n", "# Create the training and test sets\n", "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=123)\n", "\n", "# Instantiatethe XGBRegressor: xg_reg\n", "xg_reg = xgb.XGBRegressor(objective='reg:squarederror', seed=123, n_estimators=10)\n", "\n", "# Fit the regressor to the training set\n", "xg_reg.fit(X_train, y_train)\n", "\n", "# Predict the labels of the test set: preds\n", "preds = xg_reg.predict(X_test)\n", "\n", "# compute the rmse: rmse\n", "rmse = np.sqrt(mean_squared_error(y_test, preds))\n", "print(\"RMSE: %f\" % (rmse))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Linear base learners\n", "Now that you've used trees as base models in XGBoost, let's use the other kind of base model that can be used with XGBoost - a linear learner. This model, although not as commonly used in XGBoost, allows you to create a regularized linear regression using XGBoost's powerful learning API. However, because it's uncommon, you have to use XGBoost's own non-scikit-learn compatible functions to build the model, such as `xgb.train()`.\n", "\n", "In order to do this you must create the parameter dictionary that describes the kind of booster you want to use (similarly to how you created the dictionary in Chapter 1 when you used `xgb.cv()`). The key-value pair that defines the booster type (base model) you need is `\"booster\":\"gblinear\"`.\n", "\n", "Once you've created the model, you can use the `.train()` and `.predict()` methods of the model just like you've done in the past." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "RMSE: 45098.309059\n" ] } ], "source": [ "# Convert the training and testing sets into DMatrixes: DM_train, DM_test\n", "DM_train = xgb.DMatrix(data=X_train, label=y_train)\n", "DM_test = xgb.DMatrix(data=X_test, label=y_test)\n", "\n", "# Create the parameter dictionary: params\n", "params = {\"booster\":\"gblinear\", \"objective\":\"reg:squarederror\"}\n", "\n", "# Train the model: xg_reg\n", "xg_reg = xgb.train(params=params, dtrain=DM_train, num_boost_round=5)\n", "\n", "# Predict the labels of the test set: preds\n", "preds = xg_reg.predict(DM_test)\n", "\n", "# Compute and print the RMSE\n", "rmse = np.sqrt(mean_squared_error(y_test, preds))\n", "print(\"RMSE: %f\" % (rmse))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Evaluating model quality\n", "It's now time to begin evaluating model quality.\n", "\n", "Here, you will compare the RMSE and MAE of a cross-validated XGBoost model on the Ames housing data. " ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ " train-rmse-mean train-rmse-std test-rmse-mean test-rmse-std\n", "0 141767.531250 429.454591 142980.433594 1193.798509\n", "1 102832.546875 322.475723 104891.396485 1223.159762\n", "2 75872.615234 266.472656 79478.937500 1601.344539\n", "3 57245.652344 273.624748 62411.920899 2220.150028\n", "4 44401.296875 316.425549 51348.278320 2963.377786\n", "4 51348.27832\n", "Name: test-rmse-mean, dtype: float64\n" ] } ], "source": [ "# Create the DMatrix: housing_dmatrix\n", "housing_dmatrix = xgb.DMatrix(data=X, label=y)\n", "\n", "# Create the parameter dictionary: params\n", "params = {\"objective\":\"reg:squarederror\", \"max_depth\":4}\n", "\n", "# Perform cross-valdiation: cv_results\n", "cv_results = xgb.cv(dtrain=housing_dmatrix, params=params, nfold=4,\n", " num_boost_round=5, metrics='rmse', as_pandas=True, seed=123)\n", "\n", "# Print cv_results\n", "print(cv_results)\n", "\n", "# Extract and print final boosting round metric\n", "print((cv_results['test-rmse-mean']).tail(1))" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ " train-mae-mean train-mae-std test-mae-mean test-mae-std\n", "0 127343.462890 668.331024 127633.988281 2404.014089\n", "1 89770.056640 456.961189 90122.503906 2107.909888\n", "2 63580.789062 263.409865 64278.559571 1887.569709\n", "3 45633.156250 151.883825 46819.168945 1459.817731\n", "4 33587.090820 87.000638 35670.646484 1140.607452\n", "4 35670.646484\n", "Name: test-mae-mean, dtype: float64\n" ] } ], "source": [ "# Create the DMatrix: housing_dmatrix\n", "housing_dmatrix = xgb.DMatrix(data=X, label=y)\n", "\n", "# Create the parameter dictionary: params\n", "params = {\"objective\":\"reg:squarederror\", \"max_depth\":4}\n", "\n", "# Perform cross-valdiation: cv_results\n", "cv_results = xgb.cv(dtrain=housing_dmatrix, params=params, nfold=4,\n", " num_boost_round=5, metrics='mae', as_pandas=True, seed=123)\n", "\n", "# Print cv_results\n", "print(cv_results)\n", "\n", "# Extract and print final boosting round metric\n", "print((cv_results['test-mae-mean']).tail(1))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Regularization and base learners in XGBoost\n", "- Regularization in XGBoost\n", " - Regularization is a control on model complexity\n", " - Want models that are both accurate and as simple as possible\n", " - Regularization parameters in XGBoost:\n", " - Gamma - minimum loss reduction allowed for a split to occur\n", " - alpha - L1 regularization on leaf weights, larger values mean more regularization\n", " - lambda - L2 regularization on leaf weights\n", "- Base learners in XGBoost\n", " - Linear Base learner\n", " - Sum of linear terms\n", " - Boosted model is weighted sum of linear models (thus is itself linear)\n", " - Rarely used\n", " - Tree Base learner\n", " - Decision tree\n", " - Boosted model is weighted sum of decision trees (nonlinear)\n", " - Almost exclusively used in XGBoost" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Using regularization in XGBoost\n", "Having seen an example of l1 regularization in the video, you'll now vary the l2 regularization penalty - also known as `\"lambda\"` - and see its effect on overall model performance on the Ames housing dataset." ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Best rmse as a function of l2:\n", " l2 rmse\n", "0 1 52275.355468\n", "1 10 57746.062500\n", "2 100 76624.625000\n" ] } ], "source": [ "# Create the DMatrix: housing_dmatrix\n", "housing_dmatrix = xgb.DMatrix(data=X, label=y)\n", "\n", "reg_params = [1, 10, 100]\n", "\n", "# Create the initial parameter dictionary for varying l2 strength: params\n", "params = {\"objective\":\"reg:squarederror\", \"max_depth\":3}\n", "\n", "# Create an empty list for storing rmses as a function of l2 complexity\n", "rmses_l2 = []\n", "\n", "# Iterate over reg_params\n", "for reg in reg_params:\n", " # Update l2 strength\n", " params['lambda'] = reg\n", " \n", " # Pass this updated param dictionary into cv\n", " cv_results_rmse = xgb.cv(dtrain=housing_dmatrix, params=params, nfold=2,\n", " num_boost_round=5, metrics='rmse', as_pandas=True, seed=123)\n", " \n", " # Append best rmse (final round) to rmses_l2\n", " rmses_l2.append(cv_results_rmse['test-rmse-mean'].tail(1).values[0])\n", " \n", "# Loot at best rmse per l2 param\n", "print(\"Best rmse as a function of l2:\")\n", "print(pd.DataFrame(list(zip(reg_params, rmses_l2)), columns=[\"l2\", \"rmse\"]))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Visualizing individual XGBoost trees\n", "Now that you've used XGBoost to both build and evaluate regression as well as classification models, you should get a handle on how to visually explore your models. Here, you will visualize individual trees from the fully boosted model that XGBoost creates using the entire housing dataset.\n", "\n", "XGBoost has a `plot_tree()` function that makes this type of visualization easy. Once you train a model using the XGBoost learning API, you can pass it to the `plot_tree()` function along with the number of trees you want to plot using the `num_trees` argument." ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# Create the DMatrix: housing_dmatrix\n", "housing_dmatrix = xgb.DMatrix(data=X, label=y)\n", "\n", "# Create the parameters dictionary: params\n", "params = {\"objective\":'reg:squarederror', 'max_depth':2}\n", "\n", "# Train the model: xg_reg\n", "xg_reg = xgb.train(dtrain=housing_dmatrix, params=params, num_boost_round=10)\n", "\n", "# Plot the first tree\n", "fig, ax = plt.subplots(figsize=(15, 15))\n", "xgb.plot_tree(xg_reg, num_trees=0, ax=ax);\n", "\n", "# Plot the fifth tree\n", "fig, ax = plt.subplots(figsize=(15, 15))\n", "xgb.plot_tree(xg_reg, num_trees=4, ax=ax);\n", "\n", "# Plot the last tree sideways\n", "fig, ax = plt.subplots(figsize=(15, 15))\n", "xgb.plot_tree(xg_reg, rankdir=\"LR\", num_trees=9, ax=ax);" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Have a look at each of the plots. They provide insight into how the model arrived at its final decisions and what splits it made to arrive at those decisions. This allows us to identify which features are the most important in determining house price. In the next exercise, you'll learn another way of visualizing feature importances." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Visualizing feature importances: What features are most important in my dataset\n", "Another way to visualize your XGBoost models is to examine the importance of each feature column in the original dataset within the model.\n", "\n", "One simple way of doing this involves counting the number of times each feature is split on across all boosting rounds (trees) in the model, and then visualizing the result as a bar graph, with the features ordered according to how many times they appear. XGBoost has a `plot_importance()` function that allows you to do exactly this, and you'll get a chance to use it in this exercise!" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# Create the DMatrix: housing_dmatrix\n", "housing_dmatrix = xgb.DMatrix(data=X, label=y)\n", "\n", "# Create the parameter dictionary: params\n", "params = {\"objective\":\"reg:squarederror\", \"max_depth\":4}\n", "\n", "# Train the model: xg_reg\n", "xg_reg = xgb.train(dtrain=housing_dmatrix, params=params, num_boost_round=10)\n", "\n", "# Plot the feature importance\n", "xgb.plot_importance(xg_reg);\n", "plt.tight_layout()" ] } ], "metadata": { "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" } }, "nbformat": 4, "nbformat_minor": 4 }