Tensor Extraction of Latent Features (T-ELF)

T-ELF is one of the machine learning software packages developed as part of the R&D 100 winning SmartTensors AI project at Los Alamos National Laboratory (LANL). T-ELF presents an array of customizable software solutions crafted for analysis of datasets. Acting as a comprehensive toolbox, T-ELF specializes in data pre-processing, extraction of latent features, and structuring results to facilitate informed decision-making. Leveraging high-performance computing and cutting-edge GPU architectures, our toolbox is optimized for analyzing large datasets from diverse set of problems.

Central to T-ELF's core capabilities lie non-negative matrix and tensor factorization solutions for discovering multi-faceted hidden details in data, featuring automated model determination facilitating the estimation of latent factors or rank. This pivotal functionality ensures precise data modeling and the extraction of concealed patterns. Additionally, our software suite incorporates cutting-edge modules for both pre-processing and post-processing of data, tailored for diverse tasks including text mining, Natural Language Processing, and robust tools for matrix and tensor analysis and construction.

T-ELF's adaptability spans across a multitude of disciplines, positioning it as a robust AI and data analytics solution. Its proven efficacy extends across various fields such as Large-scale Text Mining, High Performance Computing, Computer Security, Applied Mathematics, Dynamic Networks and Ranking, Biology, Material Science, Medicine, Chemistry, Data Compression, Climate Studies, Relational Databases, Data Privacy, Economy, and Agriculture.

Installation

Step 1: Install Poetry to your system

This step is optional. Use Pip or Conda if Poetry is not avaiable.

Step 2: Install the Library

Option 1: Install via Poetry or Pip shell conda create --name TELF python=3.11.10 source activate TELF # or <conda activate TELF> poetry install # or <pip install .>

Option 2: Install via Conda shell git clone https://gitlab.lanl.gov/maksim/telf_internal cd telf_internal conda env create --file environment_gpu.yml # use <conda env create --file environment_cpu.yml> for CPU only conda activate TELF_conda conda develop .

Step 3: Post-installation Dependencies

Next, we need to install the optional and additional dependencies. These include optional dependencies for GPU and HPC capabilities, as well as required dependencies like the SpaCy language models. To view all available options, please run: shell python post_install.py --help Install the additional dependencies: shell python post_install.py # use the following, for example, for GPU system: <python post_install.py --gpu>

Jupyter Setup Tutorial for using the examples (Link)

Capabilities

Modules

TELF.factorization

| Method | Dense | Sparse | GPU | CPU | Multiprocessing | HPC | Description | Example | |:-------------------------:|:------------------:|:------------------:|:------------------:|:------------------:|:-------------------:|:------------------:|:----------------------------------------------------------------:|:-----------:| | NMFk | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | NMF with Automatic Model Determination | Link | | Custom NMFk | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | Use Custom NMF Functions with NMFk | Link | | TriNMFk | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | | NMF with Automatic Model Determination for Clusters and Patterns | Link | | RESCALk | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | RESCAL with Automatic Model Determination | Link | | RNMFk | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | Recommender NMFk | Link | | SymNMFk | :heavycheckmark: | | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | NMFk with Symmetric Clustering | Link | | WNMFk | :heavycheckmark: | | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | NMFk with weighting - used for recommendation system | Link | | HNMFk | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | Hierarchical NMFk | Link | | BNMFk | :heavycheckmark: | | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | Boolean NMFk | Link | | LMF | :heavycheckmark: | | :heavycheckmark: | :heavycheckmark: | | | Logistic Matrix Factorization | Link | | SPLIT | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | | Joint NMFk factorization of multiple data via SPLIT | Link | | SPLITTransfer | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | :heavycheckmark: | | Supervised transfer learning method via SPLIT and NMFk | Link |

TELF.pre_processing

| Method | Multiprocessing | HPC | Description | Example | |:----------:|:-------------------:|:-------------------:|:------------------------------------------------------------------:|:-----------:| | Vulture | :heavycheckmark: | :heavycheckmark: | Advanced text processing tool for cleaning and NLP | Link | | Beaver | :heavycheckmark: | :heavycheckmark: | Fast matrix and tensor building tool for text mining | Link | | iPenguin | :heavycheckmark: | | Online information retrieval tool for Scopus, SemanticScholar, and OSTI | Link | | Orca | :heavycheckmark: | | Duplicate author detector for text mining and information retrieval | Link | | Squirrel | | | Dataset pruning tool for documents | Link |

TELF.post_processing

| Method | Description | Example | |:----------:|:----------------------------------------------------------:|:-----------:| | Wolf | Graph centrality and ranking tool | Link | | Peacock | Data visualization and generation of actionable statistics | Link | | SeaLion | Generic report generation tool | Link | | Fox | Report generation tool for text data from NMFk using OpenAI | Link | | ArcticFox | Report generation tool for text data from HNMFk using local LLMs | Link |

TELF.applications

| Method | Description | Example | |:----------:|:--------------------------------------------------------------------:|:-----------:| | Cheetah | Fast search by keywords and phrases | Link | | Bunny | Dataset generation tool for documents and their citations/references | Link | | Penguin | Text storage tool | Link | | Lynx | Streamlit UI | Link | | Termite | Knowladge graph building tool | :soon: |

Use Cases

| Example | Description | Link | |:----------:|:--------------------------------------------------------------------:|:-----------:| | NM Law Data | Domain specific data for AI and RAG system written in our paper about New Mexico Law that uses the TELF pipeline | Link| | Full TELF Pipeline | An end-to-end pipeline demonstration, from collection to analysis | Link |

How to Cite T-ELF?

If you use T-ELF please cite.

APA: latex Eren, M., Solovyev, N., Barron, R., Bhattarai, M., Truong, D., Boureima, I., Skau, E., Rasmussen, K., & Alexandrov, B. (2023). Tensor Extraction of Latent Features (T-ELF) [Computer software]. https://doi.org/10.5281/zenodo.10257897

BibTeX: latex @software{TELF, author = {Eren, Maksim and Solovyev, Nick and Barron, Ryan and Bhattarai, Manish and Truong, Duc and Boureima, Ismael and Skau, Erik and Rasmussen, Kim and Alexandrov, Boian}, month = oct, title = {{Tensor Extraction of Latent Features (T-ELF)}}, url = {https://github.com/lanl/T-ELF}, doi = {10.5281/zenodo.10257897}, year = {2023} }

Authors

• Maksim Ekin Eren: Information Systems and Modeling Group, Los Alamos National Laboratory (Website)
• Nicholas Solovyev: Theoretical Division, Los Alamos National Laboratory
• Ryan Barron: Theoretical Division, Los Alamos National Laboratory
• Manish Bhattarai: Theoretical Division, Los Alamos National Laboratory
• Duc Truong: Theoretical Division, Los Alamos National Laboratory
• Ismael Boureima: Theoretical Division, Los Alamos National Laboratory
• Erik Skau: Computer, Computational, and Statistical Sciences Division, Los Alamos National Laboratory
• Kim Rasmussen: Theoretical Division, Los Alamos National Laboratory
• Boian S. Alexandrov: Theoretical Division, Los Alamos National Laboratory

Patents

Boian ALEXANDROV, o. S. F., New Mexico, Maksim Ekin EREN, of Sante Fe, New Mexico, Manish BHATTARAI, of Albuquerque, New Mexico, Kim Orskov RASMUSSEN of Sante Fe, New Mexico, and Charles K. NICHOLAS, of Columbia, Maryland, (“Assignor”) DATA IDENTIFICATION AND CLASSIFICATION METHOD, APPARATUS, AND SYSTEM. No. 63/472,188. Triad National Security, LLC. (June 9, 2023).

BS. Alexandrov, LB. Alexandrov, and VG. Stanev et al. 2020. Source identification by non-negative matrix factorization combined with semi-supervised clustering. US Patent S10,776,718 (2020).

Copyright Notice

© 2022. Triad National Security, LLC. All rights reserved. This program was produced under U.S. Government contract 89233218CNA000001 for Los Alamos National Laboratory (LANL), which is operated by Triad National Security, LLC for the U.S. Department of Energy/National Nuclear Security Administration. All rights in the program are reserved by Triad National Security, LLC, and the U.S. Department of Energy/National Nuclear Security Administration. The Government is granted for itself and others acting on its behalf a nonexclusive, paid-up, irrevocable worldwide license in this material to reproduce, prepare derivative works, distribute copies to the public, perform publicly and display publicly, and to permit others to do so.

LANL C Number: C22048

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

LANL HPC Installation Notes

Chicoma

```shell

replace with your own path below.

conda create --prefix= python=3.11.10 source activate # or use conda activate <...> pip install . python post_install.py --gpu --hpc-conda ```

Darwin

shell salloc -n 1 -p shared-gpu module load openmpi module load miniconda3 conda create --name TELF python=3.11.10 conda activate TELF # or <source activate TELF> pip install . python post_install.py --gpu --hpc