Xgboost plottree error: Unable to parse node: 0:[petal

Question

I'm trying to use the plot_tree as in this tutorial

I'm using the iris dataset to train the model, this is the code I have:

from numpy import loadtxt
from xgboost import XGBClassifier
from xgboost import plot_tree
import matplotlib.pyplot as plt
import shap



X,y = shap.datasets.iris()
y = pd.Series([v if v==1 else 0 for v in y]) 

model = XGBClassifier()
model.fit(X, y)
# plot single tree
plot_tree(model)
plt.show()

From there, I'm getting this error:

ValueError: Unable to parse node: 0:[petal

I don't know where to look for that since the model does the training and prediction without issues.

I'm using version '0.20.3' of sklearn

confused_zebra · Accepted Answer · 2019-10-22 12:44:21Z

2

I had the same issue and it wasn't related to the graphviz installation. In my case the problem was that some of my column names of the pandas dataFrame had white space. See also discussion on github.

When I added

df.rename(columns = lambda x: x.replace(' ', '_'), inplace=True)

to my preprocessing it solved the issue.

answered Oct 22, 2019 at 12:44

confused_zebra

1361 silver badge5 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Luis Ramon Ramirez Rodriguez Over a year ago

Yeah, was the same thing for me. Pretty werid

Vatsal Gupta · Accepted Answer · 2019-08-09 06:37:16Z

0

Make sure you have graphviz installed. Because the plot_tree of XGboost internally uses graphviz for plotting.

answered Aug 9, 2019 at 6:37

Vatsal Gupta

5213 silver badges11 bronze badges

Collectives™ on Stack Overflow

Xgboost plottree error: Unable to parse node: 0:[petal

2 Answers 2

1 Comment

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

Comments

Your Answer

Sign up or log in

Post as a guest

Related