1

When I plot my sklearn decision tree using sklearn.tree.plot_tree(), the nodes are overlapping on the deeper levels and I cannot read what is in the nodes. It is not nice to present your results. It looks like this:

overlapping nodes, want more space between them

How can I achieve having more space between them, to avoid an overlap?

I already found this question, and I already increased the fontsize to 10 and figsize to (20,30), but it leads to the tree plotted as shown in the picture. I also increased dpi to 150. This is my code:

fig = plt.figure(figsize=(20,30))
artists = sklearn.tree.plot_tree(decision_tree=clf, feature_names=feature_names, class_names=class_names, filled=True, rounded=True, fontsize=10, max_depth=4)
fig.savefig(filepath, dpi=150)
LBoss
  • 496
  • 6
  • 15

2 Answers2

0

Just increase figsize=(50,30), adjust dpi=300 and apply the code to save the image in png. This saved image should look better

fig = plt.figure(figsize=(50,30))
artists = sklearn.tree.plot_tree(decision_tree=clf, feature_names=feature_names, class_names=class_names, filled=True, rounded=True, fontsize=10, max_depth=4,dpi=300) #adjust the dpi to the parameter that fits best your output
plt.savefig('decision_tree.png')#save the tree image in png format
plt.show()
Javier Alejandro Rendon Carril
  • 129
  • 1
  • 9
  • plot_tree does not include "dpi". That is either used when creating the figure or in plt.savefig() – Aku Oct 10 '22 at 07:30
-1

You need to control the size of rendering the image. Try changing the figsize like fig = plt.figure(figsize=(50,30)). This worked for me.
You can find the original documentation in https://scikit-learn.org/stable/modules/generated/sklearn.tree.plot_tree.html.

Archana Jalaja Surendran
  • 117
  • 14
  • I appreciate your effort to answer me, but as described in the question, I already tried this and it did not work. – LBoss May 21 '21 at 08:17