SimplePPT¶
A python implementation of SimplePPT algorithm, with GPU acceleration.
Installation¶
pip install -U dyntrack
Usage¶
from sklearn.datasets import make_classification
import simpleppt
X1, Y1 = make_classification(n_features=2, n_redundant=0, n_informative=2,
n_clusters_per_class=1, n_classes=3)
SP = simpleppt.ppt(X1,Nodes=30,seed=1,progress=False,lam=10)
simpleppt.project_ppt(SP, X1, c=Y1)
Citation¶
Please cite the following paper if you use it:
Mao et al. (2015), SimplePPT: A simple principal tree algorithm
SIAM International Conference on Data Mining.
GPU dependencies (optional)¶
If you have a nvidia GPU, simpleppt can leverage CUDA computations for speedup in tree inference. The latest version of rapids framework is required (at least 0.17) it is recommanded to create a new conda environment:
conda create -n SimplePPT-gpu -c rapidsai -c nvidia -c conda-forge -c defaults \
rapids=0.19 python=3.8 cudatoolkit=11.0 -y
conda activate SimplePPT-gpu
pip install simpleppt