fixes DecisionTreeRegressor bug with sklearn 0.24

TODO.md

@@ -45,6 +45,7 @@
 
 
 ## Dashboard:
+- add reloader=None, debug=None, host=None options
 - add waitress to run options
 - add SimpleClassifierDashboard
 - add SimpleRegressionDashboard
@@ -62,6 +63,10 @@
 ## Hub:
 - automatic reloads with watchdog
 - add dashboard specific logins
+- add reloader=None, debug=None, host=None options
+- add Flask-Login support
+- add explainerhub CLI
+- add user management
 
 ### Components
 - autodetect when uuid name get rendered and issue warning
@@ -91,6 +96,7 @@
 - Add this method? : https://arxiv.org/abs/2006.04750?
 
 ## Tests:
+- add tests for InterpretML EBM (shap 0.37)
 - write tests for ExplainerHub
 - test model_output='probability' and 'raw' or 'logodds' seperately
 - write tests for explainer_methods

explainerdashboard/explainer_plots.py

@@ -2131,7 +2131,8 @@ def plotly_rf_trees(model, observation, y=None, highlight_tree=None,
             f"{highlight_tree} is out of range (0, {len(model.estimators_)})"
         colors[highlight_tree] = 'red'
 
-    if model.estimators_[0].classes_[0] is not None: #if classifier
+    if (hasattr(model.estimators_[0], "classes_") 
+        and model.estimators_[0].classes_[0] is not None): #if classifier
         preds_df = (
             pd.DataFrame({
                 'model' : range(len(model.estimators_)),