Visualizing a Decision Tree in Jupyter Notebook

Question

Is there a way to 'declump' the following tree on Jupyter Notebook? Its a simple decision tree but I do not know what is making it look collapsed. Here are the relevant code snippets and the tree itself.

%matplotlib inline
%config InlineBackend.figure_format = 'retina'
from matplotlib import pyplot as plt
plt.rcParams['figure.figsize'] = (10, 8)

import numpy as np
import pandas as pd
from sklearn.preprocessing import LabelEncoder
import collections
from sklearn.model_selection import GridSearchCV, cross_val_score
from sklearn.tree import DecisionTreeClassifier, plot_tree
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score


#some more code


# Some feature values are present in train and absent in test and vice-versa.
y = df_train['Should_Vacation_there']
df_train, df_test = intersect_features(train=df_train, test=df_test)
df_train

#training a decision tree
dt = DecisionTreeClassifier(criterion='entropy', random_state=17)
dt.fit(df_train, y);

#displaying the tree
plot_tree(dt, feature_names=df_train.columns, filled=True,
         class_names=["Should go there", "Shouldn't go there"]);

enter image description here

Check the size of the axes you are plotting the figure on. You can increase the figsize parameter in matplotlib.pyplot.figure to increase the size of the axes. Something like plt.figure(figsize=(10, 5)). — Nikhil Kumar
– Nikhil Kumar, Commented Aug 23, 2020 at 16:26
@NikhilKumar That may not solve the problem fully. Scikit-learn has a terrible visualization for decision trees. — ywbaek
– ywbaek, Commented Aug 23, 2020 at 16:28

Akshay Sehgal · Accepted Answer · 2020-08-23 22:35:33Z

The #%config InlineBackend.figure_format = 'retina' is the culprit here. Commenting it results in a well-formatted tree.

from matplotlib import pyplot as plt
import numpy as np
import pandas as pd
from sklearn.preprocessing import LabelEncoder
import collections
from sklearn.model_selection import GridSearchCV, cross_val_score
from sklearn.tree import DecisionTreeClassifier, plot_tree
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score
from sklearn.model_selection import train_test_split
from sklearn.datasets import load_iris

%matplotlib inline
#%config InlineBackend.figure_format = 'retina'

iris = load_iris()
plt.rcParams['figure.figsize'] = (10, 8)

#some more code
X_train, X_test, y_train, y_test = train_test_split(iris.data, iris.target, test_size=0.2)

# Some feature values are present in train and absent in test and vice-versa.
#y = df_train['Should_Vacation_there']
#df_train, df_test = intersect_features(train=df_train, test=df_test)
#df_train

#training a decision tree
dt = DecisionTreeClassifier(criterion='entropy', random_state=17)
dt.fit(X_train, y_train)

#displaying the tree
plot_tree(dt, feature_names=iris.feature_names, filled=True,
         class_names=iris.target_names);

aerijman · Accepted Answer · 2020-08-23 22:28:40Z

Take %config InlineBackend.figure_format = 'retina' out. With 'svg' instead, you will get great resolution.

from matplotlib import pyplot as plt
from sklearn import datasets
from sklearn.tree import DecisionTreeClassifier 
from sklearn import tree

# Prepare the data data
iris = datasets.load_iris()
X = iris.data
y = iris.target

# Fit the classifier with default hyper-parameters
clf = DecisionTreeClassifier(random_state=1234)
model = clf.fit(X, y)

1:

fig = plt.figure(figsize=(25,20))
_ = tree.plot_tree(clf, 
                   feature_names=iris.feature_names,  
                   class_names=iris.target_names,
                   filled=True)

#2

fig = plt.figure(figsize=(25,20))
_ = tree.plot_tree(clf, 
                   feature_names=iris.feature_names,  
                   class_names=iris.target_names,
                   filled=True)

Collectives™ on Stack Overflow

Visualizing a Decision Tree in Jupyter Notebook

2 Answers 2

Comments

1:

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

1:

Comments

Your Answer

Sign up or log in

Post as a guest

Related