Welcome to pyCP_APR's documentation!#

RD100

pyCP_APR is a Python library for tensor decomposition and anomaly detection that is developed as part of the R&D 100 award wining SmartTensors AI project. It is designed for the fast analysis of large datasets by accelerating computation speed using GPUs. pyCP_APR uses the CANDECOMP/PARAFAC Alternating Poisson Regression (CP-APR) tensor factorization algorithm utilizing both Numpy and PyTorch backend. While the Numpy backend can be used for the analysis of both sparse and dense tensors, PyTorch backend provides faster decomposition of large and sparse tensors on the GPU. pyCP_APR's Scikit-learn like API allows comfortable interaction with the library, and include the methods for anomaly detection via the p-values obtained from the CP-APR factorization. The anomaly detection methods via the p-values optained from CP-APR was introduced by Eren et al. in [ErenMooreAlexandrov20] using the Unified Host and Network Dataset [TKH18]. Our work follows the MATLAB Tensor Toolbox [BK06, BK08, BK+15] implementation of CP-APR [CK12].

Resources#

Installation#

Option 1: Install using pip

pip install git+https://github.com/lanl/pyCP_APR.git

Option 2: Install from source

git clone https://github.com/lanl/pyCP_APR.git
cd pyCP_APR
conda create --name pyCP_APR python=3.9
source activate pyCP_APR
pip install -e . # or <python setup.py install>

Optional Tutorial for Examples: Jupyter Setup Tutorial for using the examples (Link)

Example Usage#

from pyCP_APR import CP_APR
from pyCP_APR.datasets import load_dataset

# Load a sample tensor
data = load_dataset(name="TOY")
coords_train, nnz_train = data['train_coords'], data['train_count']
coords_test, nnz_test = data['test_coords'], data['test_count']

# CP-APR Object with PyTorch backend on a GPU. Transfer the latent factors back to Numpy arrays.
model = CP_APR(n_iters=10,
               random_state=42,
               verbose=1,
               method='torch',
               device='gpu',
               return_type='numpy')

# Train a rank 45 tensor
M = model.fit(coords=coords_train, values=nnz_train, rank=45)

# Predict the scores over the trained tensor
y_score = model.predict_scores(coords=coords_test, values=nnz_test)

How to Cite pyCP_APR?#

@MISC{Eren2021pyCPAPR,
  author = {M. E. {Eren} and J. S. {Moore} and E. {Skau} and M. {Bhattarai} and G. {Chennupati} and B. S. {Alexandrov}},
  title = {pyCP\_APR},
  year = {2021},
  publisher = {GitHub},
  journal = {GitHub repository},
  doi = {10.5281/zenodo.4840598},
  howpublished = {\url{https://github.com/lanl/pyCP\_APR}}
}


@INPROCEEDINGS{Eren2020ISI,
  author={M. E. {Eren} and J. S. {Moore} and B. S. {Alexandrov}},
  booktitle={2020 IEEE International Conference on Intelligence and Security Informatics (ISI)},
  title={Multi-Dimensional Anomalous Entity Detection via Poisson Tensor Factorization},
  year={2020},
  pages={1-6},
  doi={10.1109/ISI49825.2020.9280524}
}

Authors#

  • Maksim Ekin Eren: Advanced Research in Cyber Systems, Los Alamos National Laboratory

  • Juston S. Moore: Advanced Research in Cyber Systems, Los Alamos National Laboratory

  • Erik Skau: Computer, Computational, and Statistical Sciences Division, Los Alamos National Laboratory

  • Manish Bhattarai: Theoretical Division, Los Alamos National Laboratory

  • Gopinath Chennupati: Computer, Computational, and Statistical Sciences Division, Los Alamos National Laboratory

  • Boian S. Alexandrov: Theoretical Division, Los Alamos National Laboratory

Acknowledgments#

We thank Austin Thresher for the valuable feedback on our software design.

License#

This program is open source under the BSD-3 License. Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.

2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.

3. Neither the name of the copyright holder nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

Developer Test Suite#

Developer test suites are located under tests/ directory (located here).

Tests can be ran from this folder using python -m unittest *.

References#

[BK06]

Brett W. Bader and Tamara G. Kolda. Algorithm 862: matlab tensor classes for fast algorithm prototyping. ACM Trans. Math. Softw., 32(4):635–653, December 2006. URL: https://doi.org/10.1145/1186785.1186794, doi:10.1145/1186785.1186794.

[BK08]

Brett W. Bader and Tamara G. Kolda. Efficient matlab computations with sparse and factored tensors. SIAM Journal on Scientific Computing, 30(1):205–231, 2008. URL: https://doi.org/10.1137/060676489, arXiv:https://doi.org/10.1137/060676489, doi:10.1137/060676489.

[BK+15]

Brett W. Bader, Tamara G. Kolda, and others. Matlab tensor toolbox version 2.6. Available online, February 2015. URL: http://www.sandia.gov/~tgkolda/TensorToolbox/.

[CK12]

Eric C. Chi and Tamara G. Kolda. On tensors, sparsity, and nonnegative factorizations. SIAM Journal on Matrix Analysis and Applications, 33(4):1272–1299, December 2012. doi:10.1137/110859063.

[TKH18]

Melissa J. M. Turcotte, Alexander D. Kent, and Curtis Hash. Unified Host and Network Data Set, chapter Chapter 1, pages 1–22. World Scientific, nov 2018. URL: https://www.worldscientific.com/doi/abs/10.1142/9781786345646\_001, arXiv:https://www.worldscientific.com/doi/pdf/10.1142/9781786345646\_001, doi:10.1142/9781786345646_001.

[ErenMooreAlexandrov20]

M. E. Eren, J. S. Moore, and B. S. Alexandrov. Multi-dimensional anomalous entity detection via poisson tensor factorization. In 2020 IEEE International Conference on Intelligence and Security Informatics (ISI), volume, 1–6. 2020. doi:10.1109/ISI49825.2020.9280524.