{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Article outline\n", "\n", "1. [Intuition](#1.-Intuition)\n", "2. [Illustrating permutation importance](#2.-Illustrating-permutation-importance)\n", "3. [Sklearn Random Forest Feature Importance](#3.-Sklearn-Random-Forest-Feature-Importance)\n", "4. [Practical example](#4.-Practical-example)\n", "5. [Demo assignment](#5.-Demo-assignment)\n", "6. [Useful resources](#6.-Useful-resources)\n", "\n", "It's quite often that you want to make out the exact reasons of the algorithm outputting a particular answer. Or at the very least to find out which input features contributed most to the result. With Random Forest, you can obtain such information quite easily." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## 1. Intuition\n", "\n", "From the picture below, it is intuitively clear that, in our credit scoring problem, *Age* is much more important than *Income*. This can be formally explained using the concept of *information gain*.\n", "\n", "\n", "\n", "In the case of many decision trees or a random forest, the closer the mean position of a feature over all the trees to the root, the more significant it is for a given classification or regression problem. Gains in the splitting criterion, such as the *Gini impurity*, obtained at each optimal split in every tree is a measure of importance that is directly associated with the splitting feature. The value of this score is distinct for each feature and accumulates over all the trees.\n", "\n", "Let's go a little deeper into the details. \n", "\n", "There exist a lot of methods to assess feature importances. Leo Breinman in his works suggested to evaluate the importance of a variable by measuring decrease of accuracy of the forest when the variable is randomly permuted or decrease of impurity of a nodes where the given variable is used for splitting. The former method is often called **permutation importance**. The latter method is used in sklearn." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Permutation importance\n", "\n", "Inspired by [this](https://www.researchgate.net/publication/5231126_Conditional_Variable_Importance_for_Random_Forests) article.\n", "The average reduction in accuracy caused by a variable is determined during the calculation of the out-of-bag error. The greater the reduction in accuracy due to an exclusion or permutation of the variable, the higher its *importance score*. For this reason, variables with a greater average reduction in accuracy are generally more significant for classification.\n", "\n", "The rationale for calculating permutation importance is the following: By randomly permuting the predictor variable $X_j$, its original association with the response $Y$ is broken. When the permuted variable $X_j$, together with all the others non-permuted variables, is used the response for the out-of-bag observations, the prediction *accuracy* decreases substantially if the original $X_j$ was associated with response. Thus, as a measure of variable importance, the difference in prediction accuracy before and after permuting is used." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "More formally: denote $\\overline{\\mathfrak{B}}^{(t)}$ as the out-of-bag sample for a tree $t$, for $t\\in\\{1, ..., N\\}$ where $N$ is the number of trees in ensemble. Then the permutation importance of variable $X_j$ in tree $t$ is \n", "\n", "$${PI}^{(t)}\\left(X_j\\right)=\\frac{\\sum_{i\\in\\overline{\\mathfrak{B}}^{(t)}}I\\left(y_i=\\hat{y}_i^{(t)}\\right)}{\\left|\\overline{\\mathfrak{B}}^{(t)}\\right|}-\\frac{\\sum_{i\\in\\overline{\\mathfrak{B}}^{(t)}}I\\left(y_i=\\hat{y}_{i,\\pi_j}^{(t)}\\right)}{\\left|\\overline{\\mathfrak{B}}^{(t)}\\right|}$$ \n", "\n", "where $\\hat{y}_i^{(t)}=f^{(t)}(\\mathbf{x}_i)$ is the predicted class for observation $i$ before and $\\hat{y}_{i, \\pi_j}^{(t)}=f^{(t)}(\\mathbf{x}_{i,\\pi_j})$ is the predicted class for observation $i$ after permuting $X_j$, $\\mathbf{x}_{i,\\pi_j}=\\left(x_{i,1}, ..., x_{i,j-1},x_{\\pi_j(i),j},x_{i,j+1},...,x_{i,p}\\right)$\n", "\n", "Note that by definition ${PI}^{(t)}=0$, if variable $X_j$ isn't in tree $t$.\n", "\n", "Now, we can give the feature importance calculation for ensembles:\n", "* not normalized:\n", "$${PI}\\left(X_j\\right)=\\frac{\\sum_{t=1}^N {PI}^{(t)}(X_j)}{N}$$\n", "* normalized by the standard deviation of the differences:\n", "$$z_j=\\frac{{PI}\\left(X_j\\right)}{\\frac{\\hat{\\sigma}}{\\sqrt{N}}}$$" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## 2. Illustrating permutation importance\n", "\n", "Let's assume that we have a toy dataset with 10 instances. Target variable can be either **'N'** or **'P'**.\n", "\n", "$$\\begin{array}{c|c|c|c|c|c|c|c|c|c}\n", " \\text{Instances}, i & 1 & 2 & 3 & 4 & 5 & 6 & 7 & 8 & 9 & 10 & \\\\ \n", " \\hline\n", " y_i & N & P & P & N & N & P & N & N & N & P \\\\\n", " \\end{array}$$\n", " \n", "We build an ensemble of 5 trees $t$, for $t\\in\\{1, ..., 5\\}$. For each tree we get out-of-bag sample (denoted $\\overline{\\mathfrak{B}}^{(t)}$ above). For example for the first tree out-of-bag sample consists of instances # 2, 4, 5, and 6.\n", "\n", "$$\\begin{array}{c|c|c|c|c|c|c|c|c|c}\n", " \\text{Tree 1} & \\text{Bootstrap-sample 1} & 10 & 9 & 7 & 8 & 1 & 3 & 9 & 10 & 10 & 7\\\\\n", " \\hline\n", " \\text{Tree 2} & \\text{Bootstrap-sample 2} & 4 & 8 & 5 & 8 & 3 & 9 & 2 & 6 & 1 & 6\\\\\n", " \\hline\n", " \\text{Tree 3} & \\text{Bootstrap-sample 3} & 6 & 2 & 6 & 10 & 2 & 10 & 3 & 6 & 5 & 1\\\\\n", " \\hline\n", " \\text{Tree 4} & \\text{Bootstrap-sample 4} & 6 & 7 & 8 & 10 & 6 & 10 & 9 & 10 & 8 & 2\\\\\n", " \\hline\n", " \\text{Tree 5} & \\text{Bootstrap-sample 5} & 5 & 8 & 1 & 8 & 5 & 7 & 10 & 1 & 10 & 9\\\\\n", " \\end{array}$$\n", " \n", "Thus, out-of-bag samples for each tree $t$ are\n", "\n", "$$\\begin{array}{c|cccc}\n", " \\text{Tree}, t & \\overline{\\mathfrak{B}}^{(t)} \\\\\n", " \\hline\n", " \\text{Tree 1} & 2 & 4 & 5 & 6\\\\\n", " \\hline\n", " \\text{Tree 2} & 7 & 10\\\\\n", " \\hline\n", " \\text{Tree 3} & 4 & 7 & 8 & 9\\\\\n", " \\hline\n", " \\text{Tree 4} & 1 & 3 & 4 & 5\\\\\n", " \\hline\n", " \\text{Tree 5} & 2 & 3 & 4 & 6\\\\\n", " \\hline\n", " \\end{array}$$" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Suppose that we have four features $X_j$, $j\\in\\{1, 2, 3, 4\\}$ and we'd like to compute _permutation importance_ for $X_2$. First, for each out-of-bag sample we compute _accuracy_ of the model before and after permutation of the values of $X_2$.\n", "\n", "For instance, before permutation for $\\overline{\\mathfrak{B}}^{(1)}$ we have\n", "\n", "$$\\begin{array}{c|cccc|cc|c}\n", " & X_1 & \\color{red}{X_2} & X_3 & X_4 & y_i & \\hat{y}_i & I\\left(y_i=\\hat{y}_i\\right)\\\\\n", " \\hline\n", " \\textbf{2} & 1 & \\color{red}2 & 11 & 101 & \\textbf{P} & \\textbf{P} & 1\\\\\n", " \\hline\n", " \\textbf{4} & 2 & \\color{red}3 & 12 & 102 & \\textbf{N} & \\textbf{P} & 0\\\\\n", " \\hline\n", " \\textbf{5} & 3 & \\color{red}5 & 13 & 103 & \\textbf{N} & \\textbf{N} & 1\\\\\n", " \\hline\n", " \\textbf{6} & 4 & \\color{red}7 & 14 & 104 & \\textbf{P} & \\textbf{P} & 1\\\\\n", " \\end{array}$$\n", " \n", "Thus, the accuracy before permutation is $3/4=0.75$.\n", " \n", "After permutation for $\\overline{\\mathfrak{B}}^{(1)}$ we have\n", "\n", "$$\\begin{array}{c|cccc|cc|c}\n", " & X_1 & \\color{red}{X_2} & X_3 & X_4 & y_i & \\hat{y}_i & I\\left(y_i=\\hat{y}_i\\right)\\\\\n", " \\hline\n", " \\textbf{2} & 1 & \\color{red}5 & 11 & 101 & \\textbf{P} & \\textbf{P} & 0\\\\\n", " \\hline\n", " \\textbf{4} & 2 & \\color{red}7 & 12 & 102 & \\textbf{N} & \\textbf{P} & 0\\\\\n", " \\hline\n", " \\textbf{5} & 3 & \\color{red}2 & 13 & 103 & \\textbf{N} & \\textbf{N} & 1\\\\\n", " \\hline\n", " \\textbf{6} & 4 & \\color{red}3 & 14 & 104 & \\textbf{P} & \\textbf{P} & 1\\\\\n", " \\end{array}$$\n", " \n", "The accuracy after permutation is $2/4=0.50$.\n", "\n", "Then the difference between accuracies is computed.\n", "\n", "The above mentioned steps are to be done for each out-of-bag sample $\\overline{\\mathfrak{B}}^{(t)}$. To get not normalized _permutation importance_ we sum all computed differences and divide by the number of trees. Normalization is done by dividing _not normalized permutation importance_ by standard error." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## 3. Sklearn Random Forest Feature Importance\n", "\n", "Inspired by [this](https://medium.com/@srnghn/the-mathematics-of-decision-trees-random-forest-and-feature-importance-in-scikit-learn-and-spark-f2861df67e3) article.\n", "Sklearn library uses another approach to determine feature importance. The rationale for that method is that the more gain in information the node (with splitting feature $X_j$) provides, the higher its importance.\n", "\n", "The average reduction in the Gini impurity – or MSE for regression – represents the contribution of each feature to the homogeneity of nodes and leaves in the resulting Random Forest model. Each time a selected feature is used for splitting, the Gini impurity of the child nodes is calculated and compared with that of the original node.\n", "\n", "Gini impurity is a score of homogeneity with the range from 0 (homogeneous) to 1 (heterogeneous). The changes in the value of the splitting criterion are accumulated for each feature and normalized at the end of the calculation. A higher reduction in the Gini impurity signals that splitting results by this feature results in nodes with higher purity.\n", "\n", "The algorithm of obtaining feature importance may be represented with the following sequence of steps:\n", "\n", "1. For each tree $t$ in ensemble $t\\in\\{1,...,N\\}$:\n", "\n", " 1.1. for each node $i$ calculate the reduction in impurity (or MSE, or entropy) as ${RI}_i^{(t)}=w_i^{(t)}\\cdot I_i^{(t)} - w_{LEFT_i}^{(t)}\\cdot I_{LEFT_i}^{(t)}-w_{RIGHT_i}^{(t)}\\cdot I_{RIGHT_i}^{(t)}$, where:\n", " - $w_i^{(t)}$, $w_{LEFT_i}^{(t)}$, and $w_{RIGHT_i}^{(t)}$ are respectively weighted number of samples reaching node $i$ in tree $t$, and its left $LEFT_i$ and right $RIGHT_i$ children\n", " - $I_i^{(t)}$, $I_{LEFT_i}^{(t)}$, $I_{RIGHT_i}^{(t)}$ are impurity of the nodes. For leaves ${RI}_i^{(t)}$ is equal to 0.\n", "\n", " 1.2. for each feature $j$ calculate its importance in that particular tree as\n", " \n", "$${FI}_j^{(t)}=\\frac{\\sum_{i:\\text{node }i\\text{ splits on feature } j}{RI}_i^{(t)}}{\\sum_{i\\in\\text{all nodes}}{RI}_i^{(t)}}$$\n", "\n", " That means that in numerator we sum the reduction in impurity only in those nodes where feature $j$ is situated.\n", "2. Calculate the average feature importances over all trees in ensemble:\n", "\n", "$${FI}_j=\\frac{\\sum_{t=1}^N {FI}_j^{(t)}}{N}$$" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Those are pretty confusing formulas so let's demonstrate each step with the Iris Dataset." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import numpy as np\n", "import pandas as pd\n", "\n", "from sklearn.datasets import load_iris\n", "import matplotlib.pyplot as plt\n", "import seaborn as sns\n", "\n", "iris = load_iris()\n", "data = iris['data']\n", "target = iris['target']" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
" ], "text/plain": [ " sepal length (cm) sepal width (cm) petal length (cm) petal width (cm)\n", "0 5.1 3.5 1.4 0.2\n", "1 4.9 3.0 1.4 0.2\n", "2 4.7 3.2 1.3 0.2\n", "3 4.6 3.1 1.5 0.2\n", "4 5.0 3.6 1.4 0.2" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data = pd.DataFrame(data, columns=iris['feature_names'])\n", "data.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Since our aim is just to demonstrate the sequence of steps in calculating feature importances we'll transform the target variable as for classifying Iris Virginica One-To-All." ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "target = pd.Series(target).map({0: 0, 1: 0, 2: 1})" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Creating Random Forest. For reproducibility, we set random_state=17. For the sake of simplicity we set the number of trees to 3 and limit the depth of trees in ensemble to be not greater than 3." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "from sklearn.ensemble import RandomForestClassifier\n", "rfc = RandomForestClassifier(n_estimators=3, max_depth=3, random_state=17)\n", "rfc.fit(data, target);" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "After fitting list of all the trees are stored in estimators_ property." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "tree_list = rfc.estimators_" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Visualizing trees" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "