graph.write_pdf("iris.pdf") AttributeError: 'list' object has no attribute 'write_pdf'

I think you are using newer version of python. Please try with pydotplus.

import pydotplus
...
graph = pydotplus.graph_from_dot_data(dot_data.getvalue())
graph.write_pdf("iris.pdf")

This should do it.

I install scikit-learn via conda and all of about not work.
Firstly, I have to install libtool

brew install libtool --universal

Then I follow this sklearn guide
Then change the python file to this code

clf = clf.fit(train_data, train_target)
tree.export_graphviz(clf,out_file='tree.dot') 

Finally convert to png in terminal

dot -Tpng tree.dot -o tree.png

@Alex Sokolov, for my case in window, i downloaded and install / unzip the following to a folder then setup the PATH in Windows environment variables. re-run the py code works for me. hope is helpful to you.

I had exactly the same issue. Turned out that I hadn’t installed graphviz. Once i did that it started to work.

I tried the previous answers and still got a error when running the script Therefore,
I just used pydotplus

import pydotplus

and install the “graphviz” by using:

sudo apt-get install graphviz

Then it worked for me, and I added

graph = pydotplus.graph_from_dot_data(dot_data.getvalue())
graph.write_pdf("iris.pdf")

Thanks to the previous contributors.

I hope this helps, I was having a similar issue. I decided not to use pydot / pydotplus, but rather graphviz. I modified (barely) the code and it works wonders! 🙂

# 2. Train classifier
# Testing Data
# Examples used to "test" the classifier's accuracy
# Not part of the training data
import numpy as np
from sklearn.datasets import load_iris
from sklearn import tree
iris = load_iris()
test_idx = [0, 50, 100] # Grabs one example of each flower for testing data (in the data set it so happens to be that
                        # each flower begins at 0, 50, and 100

# training data
train_target = np.delete(iris.target, test_idx)     # Delete all but 3 for training target data
train_data = np.delete(iris.data, test_idx, axis=0) # Delete all but 3 for training data

# testing data
test_target = iris.target[test_idx] # Get testing target data
test_data = iris.data[test_idx]     # Get testing data

# create decision tree classifier and train in it on the testing data
clf = tree.DecisionTreeClassifier()
clf.fit(train_data, train_target)

# Predict label for new flower
print(test_target)
print(clf.predict(test_data))

# Visualize the tree
from sklearn.externals.six import StringIO
import graphviz
dot_data = StringIO()
tree.export_graphviz(clf,
        out_file=dot_data,
        feature_names=iris.feature_names,
        class_names=iris.target_names,
        filled=True, rounded=True,
        impurity=False)
graph = graphviz.Source(dot_data.getvalue())
graph.render("iris.pdf", view=True)

It works as the following on Python3.7 but don’t forget to install pydot using Anaconda prompt:

   from sklearn.externals.six import StringIO
   import pydot

   # viz code
   dot_data = StringIO()
   tree.export_graphviz(clf, out_file=dot_data, feature_names=iris.feature_names,
                 class_names=iris.target_names, filled=True, rounded=True,
                 impurity=False)
   graph = pydot.graph_from_dot_data(dot_data.getvalue())
   graph[0].write_pdf('iris.pdf')

I use Anaconda. Here’s what worked for me:
run from terminal:

conda install python-graphviz
conda install pydot     ## don't forget this <-----------------

Then run

clf = clf.fit(train_data, train_target)
tree.export_graphviz(clf,out_file='tree.dot')

Then from the terminal:

dot -Tpng tree.dot -o tree.png

To add all graphs for the number of your n_estimators you can do:

for i in range(0, n):  #n is your n_estimators number
    dot_data = StringIO()
    tree.export_graphviz(clf.estimators_[i], out_file=dot_data, feature_names=iris.feature_names,
                        class_names=iris.target_names, filled=True, rounded=True,
                        impurity=False)
    graph = pydotplus.graph_from_dot_data(dot_data.getvalue())
    graph.write_pdf("iris%s.pdf"%i)

you could also switch the line

graph = pydotplus.graph_from_dot_data(dot_data.getvalue())

for this one

(graph,) = pydot.graph_from_dot_data(dot_data.getvalue())
graph.write_pdf("iris.pdf")

and it would still work.