add treegamregressor

Author: csinva

docs/algebraic/gam.html

@@ -29,14 +29,14 @@
 <section>
 <h2 class="section-title" id="header-submodules">Sub-modules</h2>
 <dl>
-<dt><code class="name"><a title="imodels.algebraic.gam" href="gam.html">imodels.algebraic.gam</a></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
 <dt><code class="name"><a title="imodels.algebraic.slim" href="slim.html">imodels.algebraic.slim</a></code></dt>
 <dd>
 <div class="desc"><p>Wrapper for sparse, integer linear models …</p></div>
 </dd>
+<dt><code class="name"><a title="imodels.algebraic.tree_gam" href="tree_gam.html">imodels.algebraic.tree_gam</a></code></dt>
+<dd>
+<div class="desc"></div>
+</dd>
 </dl>
 </section>
 <section>
@@ -65,8 +65,8 @@ <h1>Index 🔍</h1>
 </li>
 <li><h3><a href="#header-submodules">Sub-modules</a></h3>
 <ul>
-<li><code><a title="imodels.algebraic.gam" href="gam.html">imodels.algebraic.gam</a></code></li>
 <li><code><a title="imodels.algebraic.slim" href="slim.html">imodels.algebraic.slim</a></code></li>
+<li><code><a title="imodels.algebraic.tree_gam" href="tree_gam.html">imodels.algebraic.tree_gam</a></code></li>
 </ul>
 </li>
 </ul>

docs/algebraic/index.html

@@ -158,7 +158,7 @@ <h3 id="supported-models">Supported models</h3>
 </tr>
 <tr>
 <td style="text-align: left;">Tree GAM</td>
-<td><a href="https://csinva.io/imodels/algebraic/gam.html">🗂️</a>, <a href="https://github.com/interpretml/interpret">🔗</a>, <a href="https://dl.acm.org/doi/abs/10.1145/2339530.2339556">📄</a></td>
+<td><a href="https://csinva.io/imodels/algebraic/tree_gam.html">🗂️</a>, <a href="https://github.com/interpretml/interpret">🔗</a>, <a href="https://dl.acm.org/doi/abs/10.1145/2339530.2339556">📄</a></td>
 <td>Generalized additive model fit with short boosted trees</td>
 </tr>
 <tr>
@@ -363,6 +363,12 @@ <h2 id="support-for-different-tasks">Support for different tasks</h2>
 <td>Requires extra dependencies for speed</td>
 </tr>
 <tr>
+<td style="text-align: left;">Tree GAM</td>
+<td style="text-align: center;"><a href="https://csinva.io/imodels/algebraic/tree_gam.html">TreeGAMClassifier</a></td>
+<td style="text-align: center;"><a href="https://csinva.io/imodels/algebraic/tree_gam.html">TreeGAMRegressor</a></td>
+<td></td>
+</tr>
+<tr>
 <td style="text-align: left;">Greedy tree sums (FIGS)</td>
 <td style="text-align: center;"><a href="https://csinva.io/imodels/tree/figs.html#imodels.tree.figs.FIGSClassifier">FIGSClassifier</a></td>
 <td style="text-align: center;"><a href="https://csinva.io/imodels/tree/figs.html#imodels.tree.figs.FIGSRegressor">FIGSRegressor</a></td>
@@ -444,6 +450,9 @@ <h3 id="hierarchical-shrinkage-post-hoc-regularization-for-tree-based-methods">H
 <p align="center">
 <i><b>HS Example.</b> HS applies post-hoc regularization to any decision tree by shrinking each node towards its parent.</i>
 </p>
+<h3 id="mdi-a-flexible-random-forest-based-feature-importance-framework">MDI+: A Flexible Random Forest-Based Feature Importance Framework</h3>
+<p><a href="https://arxiv.org/pdf/2307.01932.pdf">📄 Paper</a>, <a href="https://scholar.google.com/scholar?hl=en&amp;as_sdt=0%2C23&amp;q=MDI%2B%3A+A+Flexible+Random+Forest-Based+Feature+Importance+Framework&amp;btnG=#d=gs_cit&amp;t=1690399844081&amp;u=%2Fscholar%3Fq%3Dinfo%3Axc0LcHXE_lUJ%3Ascholar.google.com%2F%26output%3Dcite%26scirp%3D0%26hl%3Den">📌 Citation</a></p>
+<p>MDI+ is a novel feature importance framework, which generalizes the popular mean decrease in impurity (MDI) importance score for random forests. At its core, MDI+ expands upon a recently discovered connection between linear regression and decision trees. In doing so, MDI+ enables practitioners to (1) tailor the feature importance computation to the data/problem structure and (2) incorporate additional features or knowledge to mitigate known biases of decision trees. In both real data case studies and extensive real-data-inspired simulations, MDI+ outperforms commonly used feature importance measures (e.g., MDI, permutation-based scores, and TreeSHAP) by substantional margins.</p>
 <h2 id="references">References</h2>
 <details>
 <summary>Readings</summary>
@@ -511,7 +520,7 @@ <h2 id="references">References</h2>
 # Github repo available [here](https://github.com/csinva/imodels)
 
 from .algebraic.slim import SLIMRegressor, SLIMClassifier
-from .algebraic.gam import TreeGAMClassifier
+from .algebraic.tree_gam import TreeGAMClassifier, TreeGAMRegressor
 from .discretization.discretizer import RFDiscretizer, BasicDiscretizer
 from .discretization.mdlp import MDLPDiscretizer, BRLDiscretizer
 from .experimental.bartpy import BART
@@ -529,27 +538,64 @@ <h2 id="references">References</h2>
 from .rule_set.skope_rules import SkopeRulesClassifier
 from .rule_set.slipper import SlipperClassifier
 from .tree.c45_tree.c45_tree import C45TreeClassifier
-from .tree.cart_ccp import DecisionTreeCCPClassifier, DecisionTreeCCPRegressor, HSDecisionTreeCCPClassifierCV, \
-    HSDecisionTreeCCPRegressorCV
+from .tree.cart_ccp import (
+    DecisionTreeCCPClassifier,
+    DecisionTreeCCPRegressor,
+    HSDecisionTreeCCPClassifierCV,
+    HSDecisionTreeCCPRegressorCV,
+)
+
 # from .tree.iterative_random_forest.iterative_random_forest import IRFClassifier
 # from .tree.optimal_classification_tree import OptimalTreeModel
 from .tree.cart_wrapper import GreedyTreeClassifier, GreedyTreeRegressor
 from .tree.figs import FIGSRegressor, FIGSClassifier, FIGSRegressorCV, FIGSClassifierCV
 from .tree.gosdt.pygosdt import OptimalTreeClassifier
-from .tree.gosdt.pygosdt_shrinkage import HSOptimalTreeClassifier, HSOptimalTreeClassifierCV
-from .tree.hierarchical_shrinkage import HSTreeRegressor, HSTreeClassifier, HSTreeRegressorCV, HSTreeClassifierCV
+from .tree.gosdt.pygosdt_shrinkage import (
+    HSOptimalTreeClassifier,
+    HSOptimalTreeClassifierCV,
+)
+from .tree.hierarchical_shrinkage import (
+    HSTreeRegressor,
+    HSTreeClassifier,
+    HSTreeRegressorCV,
+    HSTreeClassifierCV,
+)
 from .tree.tao import TaoTreeClassifier, TaoTreeRegressor
 from .util.data_util import get_clean_dataset
 from .util.distillation import DistilledRegressor
 from .util.explain_errors import explain_classification_errors
 
-CLASSIFIERS = [BayesianRuleListClassifier, GreedyRuleListClassifier, SkopeRulesClassifier,
-               BoostedRulesClassifier, SLIMClassifier, SlipperClassifier, BayesianRuleSetClassifier,
-               C45TreeClassifier, OptimalTreeClassifier, OptimalRuleListClassifier, OneRClassifier,
-               SlipperClassifier, RuleFitClassifier, TaoTreeClassifier,
-               FIGSClassifier, HSTreeClassifier, HSTreeClassifierCV]  # , IRFClassifier
-REGRESSORS = [RuleFitRegressor, SLIMRegressor, GreedyTreeClassifier, FIGSRegressor,
-              TaoTreeRegressor, HSTreeRegressor, HSTreeRegressorCV, BART]
+CLASSIFIERS = [
+    BayesianRuleListClassifier,
+    GreedyRuleListClassifier,
+    SkopeRulesClassifier,
+    BoostedRulesClassifier,
+    SLIMClassifier,
+    SlipperClassifier,
+    BayesianRuleSetClassifier,
+    C45TreeClassifier,
+    OptimalTreeClassifier,
+    OptimalRuleListClassifier,
+    OneRClassifier,
+    SlipperClassifier,
+    RuleFitClassifier,
+    TaoTreeClassifier,
+    TreeGAMClassifier,
+    FIGSClassifier,
+    HSTreeClassifier,
+    HSTreeClassifierCV,
+]  # , IRFClassifier
+REGRESSORS = [
+    RuleFitRegressor,
+    SLIMRegressor,
+    GreedyTreeClassifier,
+    FIGSRegressor,
+    TaoTreeRegressor,
+    TreeGAMRegressor,
+    HSTreeRegressor,
+    HSTreeRegressorCV,
+    BART,
+]
 ESTIMATORS = CLASSIFIERS + REGRESSORS
 DISCRETIZERS = [RFDiscretizer, BasicDiscretizer, MDLPDiscretizer, BRLDiscretizer]</code></pre>
 </details>
@@ -609,7 +655,14 @@ <h1>Index 🔍</h1>
 <li><a href="#support-for-different-tasks">Support for different tasks</a><ul>
 <li><a href="#extras">Extras</a></li>
 </ul>
-</li><li><a href="#references">References</a></li>
+</li>
+<li><a href="#our-favorite-models">Our favorite models</a><ul>
+<li><a href="#figs-fast-interpretable-greedy-tree-sums">FIGS: Fast interpretable greedy-tree sums</a></li>
+<li><a href="#hierarchical-shrinkage-post-hoc-regularization-for-tree-based-methods">Hierarchical shrinkage: post-hoc regularization for tree-based methods</a></li>
+<li><a href="#mdi-a-flexible-random-forest-based-feature-importance-framework">MDI+: A Flexible Random Forest-Based Feature Importance Framework</a></li>
+</ul>
+</li>
+<li><a href="#references">References</a></li>
 </ul>
 </div>
 <ul id="index">

docs/algebraic/tree_gam.html

@@ -401,7 +401,10 @@
             # require the tree to have more than 1 node, otherwise just leave importance_data_tree as zeros
             if 1 < next(node_counter):
                 tree_samples = _importances(tree_)
-                importance_data_tree /= tree_samples
+                if tree_samples != 0:
+                    importance_data_tree /= tree_samples
+                else:
+                    importance_data_tree = 0
 
             importance_data.append(importance_data_tree)
 
@@ -1015,7 +1018,10 @@ <h2 id="params">Params</h2>
             # require the tree to have more than 1 node, otherwise just leave importance_data_tree as zeros
             if 1 &lt; next(node_counter):
                 tree_samples = _importances(tree_)
-                importance_data_tree /= tree_samples
+                if tree_samples != 0:
+                    importance_data_tree /= tree_samples
+                else:
+                    importance_data_tree = 0
 
             importance_data.append(importance_data_tree)
 
@@ -1410,7 +1416,10 @@ <h3>Methods</h3>
         # require the tree to have more than 1 node, otherwise just leave importance_data_tree as zeros
         if 1 &lt; next(node_counter):
             tree_samples = _importances(tree_)
-            importance_data_tree /= tree_samples
+            if tree_samples != 0:
+                importance_data_tree /= tree_samples
+            else:
+                importance_data_tree = 0
 
         importance_data.append(importance_data_tree)