Visualising the decision tree in sklearn

Question

When I want to visualise the tree I got this error.

I have shown the required libraries imported. Is there expected reason with jupiter-notebook ?

from sklearn import tree
import matplotlib.pyplot
from sklearn.tree import DecisionTreeClassifier
from sklearn.model_selection import train_test_split
from sklearn.datasets import load_breast_cancer
cancer=load_breast_cancer()
x=cancer.data
y=cancer.target
clf=DecisionTreeClassifier(max_depth=1000)
x_train,x_test,y_train,y_test=train_test_split(x,y)
clf=clf.fit(x_train,y_train)
tree.plot_tree(clf.fit(x_train,y_train))

AttributeError: module 'sklearn.tree' has no attribute 'plot_tree'

check it out here: https://scikit-learn.org/stable/modules/tree.html — PV8, Sep 06 '19 at 07:48
make sure your `matplotlib` version is >= 1.5 and try to save your fit to an object before putting it in the plot function — Anna Yashina, Sep 06 '19 at 10:39
`plot_tree` is new in version 0.21. Maybe check your scikit-learn version — Josmoor98, Sep 06 '19 at 11:15

Anna Yashina · Answer 1 · 2019-09-06T10:56:56.710

8

I assigned the tree to an object and added plt.show(). This works for me.

%matplotlib inline
from sklearn import tree
import matplotlib.pyplot as plt
from sklearn.tree import DecisionTreeClassifier
from sklearn.model_selection import train_test_split
from sklearn.datasets import load_breast_cancer
cancer = load_breast_cancer()
x = cancer.data
y = cancer.target
clf = DecisionTreeClassifier(max_depth = 1000)
x_train,x_test,y_train,y_test = train_test_split(x,y)

fig = clf.fit(x_train,y_train)
tree.plot_tree(fig)
plt.show()

But I recommend using graphviz, it's much more flexible.

edited Sep 06 '19 at 10:56

answered Sep 06 '19 at 10:47

Anna Yashina

514
8
18

1

current version is 0.21.3 you can upgrade running `!pip install -U scikit-learn` or `conda upgrade scikit-learn` if you are using anaconda – Anna Yashina Sep 06 '19 at 11:37
Thank you so much, I have got it after upgrading – Roshan Sep 06 '19 at 12:01
3

This is not an answer. You are just saying "I don't have this problem". – Nicolas Gervais Mar 15 '20 at 18:40

score 4 · Answer 2 · answered Dec 03 '19 at 23:35

4

upgrade sklearn package:

pip install --upgrade sklearn

answered Dec 03 '19 at 23:35

Mohammad Nazari

2,535
1
18
29

score 2 · Answer 3 · answered Mar 15 '20 at 18:39

This is because plot_tree is new is sklearn version 0.21, as indicated in the documentation. Check if you have a version sufficient by running this:

import sklearn

print(sklearn.__version__)

assert float(sklearn.__version__[2:]) >= 21, 'sklearn version insufficient.'

If you get an error message, you need to update sklearn

pip install --upgrade sklearn

score 1 · Answer 4 · answered Jul 16 '21 at 20:04

Because plot_tree is defined after sklearn version 0.21

For checking Version Open any python idle Running below program.

import sklearn
print (sklearn.__version__)

If the version shows less than 0.21 then you need to upgrade the sklearn library.

Open Anaconda prompt and write below command

pip install --upgrade scikit-learn

Visualising the decision tree in sklearn

4 Answers4