mirror of https://github.com/Doctorado-ML/STree.git synced 2025-08-15 15:36:00 +00:00

Go to file

Fix mistake in computing multiclass node belief

Set default criterion for split to entropy instead of gini
Set default max_iter to 1e5 instead of 1e3
change up-down criterion to match SVC multiclass
Fix impurity method of splitting nodes
Update jupyter Notebooks

2020-11-01 17:37:17 +01:00

notebooks

Fix mistake in computing multiclass node belief

2020-11-01 17:37:17 +01:00

stree

Fix mistake in computing multiclass node belief

2020-11-01 17:37:17 +01:00

.coveragerc

better solution to the sklearn bagging problem

2020-06-26 11:22:45 +02:00

.gitignore

Get only 3 sets for best split

2020-06-27 18:29:40 +02:00

codecov.yml

#6 - Update tests and codecov conf

2020-06-11 13:45:24 +02:00

example.png

Implement grapher and notebook to test it

2020-05-22 19:42:13 +02:00

LICENSE

Initial commit

2020-05-09 23:01:30 +02:00

main.py

#2 Add predict and score support

2020-06-14 14:00:21 +02:00

pyproject.toml

#4 Add code coverage & codacy badge

2020-06-06 03:04:18 +02:00

README.md

#3 Add sample_weights to score, update notebooks

2020-06-09 01:46:38 +02:00

requirements.txt

Fix problem in _min_distance

2020-06-12 00:50:25 +02:00

setup.py

Fix mistake in computing multiclass node belief

2020-11-01 17:37:17 +01:00

README.md

Stree

Oblique Tree classifier based on SVM nodes. The nodes are built and splitted with sklearn SVC models. Stree is a sklearn estimator and can be integrated in pipelines, grid searches, etc.

Installation

pip install git+https://github.com/doctorado-ml/stree

Examples

Jupyter notebooks

Benchmark
Benchmark
Test features
Adaboost
Gridsearch
Test Graphics

Command line

python main.py

Tests

python -m unittest -v stree.tests