supertree 1.0.0

Visualize decision tree in Python

supertree - Interactive Decision Tree Visualization

supertree is a Python package designed to visualize decision trees in an interactive and user-friendly way within Jupyter Notebooks, Jupyter Lab, Google Colab, and any other notebooks that support HTML rendering. With this tool, you can not only display decision trees, but also interact with them directly within your notebook environment. Key features include:

• ability to zoom and pan through large trees,
• collapse and expand selected nodes,
• explore the structure of the tree in an intuitive and visually appealing manner.

Examples

Decision Tree classifier on iris data

from sklearn.tree import DecisionTreeClassifier
from sklearn.datasets import load_iris
from supertree import SuperTree # <- import supertree :)

# Load the iris dataset
iris = load_iris()
X, y = iris.data, iris.target

# Train model
model = DecisionTreeClassifier()
model.fit(X, y)

# Initialize supertree
super_tree = SuperTree(model, X, y, iris.feature_names, iris.target_names, color_palette=1)

# show tree in your notebook
super_tree.show_tree()

Random Forest Regressor Example

from sklearn.ensemble import RandomForestRegressor
from sklearn.datasets import load_diabetes
from supertree import SuperTree  # <- import supertree :)

# Load the diabetes dataset
diabetes = load_diabetes()
X = diabetes.data
y = diabetes.target

# Train model
model = RandomForestRegressor(n_estimators=100, max_depth=3, random_state=42)
model.fit(X, y)

# Initialize supertree
super_tree = SuperTree(model,X, y)
# show tree with index 2 in your notebook
super_tree.show_tree(2)

There are more code snippets in the examples directory.

Instalation

You can install SuperTree package using pip:

pip install supertree

Conda support coming soon.

JavaScript Source

The interactive frontend is authored in:

• supertree/js/src/
• supertree/js/dependencies/d3vs.js

The packaged runtime still ships:

• supertree/js/script.js as the generated readable bundle
• supertree/js/supertree.min.js as the generated packaged runtime

To rebuild both after editing the source files, run:

./scripts/build_js.sh

Maintainer notes and a frontend smoke checklist are in docs/frontend_development.md.

Supported Libraries

• scikit-learn (sklearn)
• LightGBM
• XGBoost
• ONNX:

Supported Algorithms

The package is compatible with a wide range of classifiers and regressors from these libraries, specifically:

Scikit-learn

• DecisionTreeClassifier
• ExtraTreeClassifier
• ExtraTreesClassifier
• RandomForestClassifier
• GradientBoostingClassifier
• HistGradientBoostingClassifier
• DecisionTreeRegressor
• ExtraTreeRegressor
• ExtraTreesRegressor
• RandomForestRegressor
• GradientBoostingRegressor
• HistGradientBoostingRegressor

LightGBM

• LGBMClassifier
• LGBMRegressor
• Booster

XGBoost

• XGBClassifier
• XGBRFClassifier
• XGBRegressor
• XGBRFRegressor
• Booster

If we do not support the model you want to use, please let us know.

Features

See all the details

Zoom

Fullscreen in Jupyter

Depth change

Color change

Navigate in forest

Show specific sample path

Save tree to svg

Links sample visualization

Showing the path to the leaf

Check this features in example directory :)

Articles

• Visualize decision tree from scikit-learn package
• 4 ways to vizualize decision tree from LightGBM
• How to visualize decision tree from Xgboost

Support

If you encounter any issues, find a bug, or have a feature request, we would love to hear from you! Please don't hesitate to reach out to us at supertree/issues. We are committed to improving this package and appreciate any feedback or suggestions you may have.

License

supertree is open source under the Apache License 2.0.