Plot trees for a Random Forest in Python with Scikit-Learn

Question

I want to plot a decision tree of a random forest. So, i create the following code:

clf = RandomForestClassifier(n_estimators=100)
import pydotplus
import six
from sklearn import tree
dotfile = six.StringIO()
i_tree = 0
for tree_in_forest in clf.estimators_:
if (i_tree <1):        
    tree.export_graphviz(tree_in_forest, out_file=dotfile)
    pydotplus.graph_from_dot_data(dotfile.getvalue()).write_png('dtree'+ str(i_tree) +'.png')
    i_tree = i_tree + 1

But it doesn't generate anything.. Have you an idea how to plot a decision tree from random forest ?

Thank you,

Answer 1

Assuming your Random Forest model is already fitted, first you should first import the export_graphviz function:

from sklearn.tree import export_graphviz

In your for cycle you could do the following to generate the dot file

export_graphviz(tree_in_forest,
                feature_names=X.columns,
                filled=True,
                rounded=True)

The next line generates a png file

os.system('dot -Tpng tree.dot -o tree.png')