The aim is to have a pip installable library which implements as many of the "cool" / "latest" / "bleeding edge" algorithms in ML as possible.
pip install bleedml
Most of the algorithms are meant as a drop in replacement for sklearn estimators.
from bleedml.classifiers import CascadeForest
from sklearn.model_selection import cross_val_score
from sklearn.datasets import load_iris
X, y = load_iris(return_X_y=True)
est = CascadeForest()
scores = cross_val_score(est, X, y, cv=3, scoring='accuracy')
print(scores.mean())