time-series-nmf: Non-negative Matrix Factorization for Time Series
| Main Authors: | Valentina Staneva, Wu-Jung Lee |
|---|---|
| Format: | info software Journal |
| Terbitan: |
, 2020
|
| Subjects: | |
| Online Access: |
https://zenodo.org/record/3906891 |
Daftar Isi:
- time-series-nmf is a Python package implementing non-negative matrix factorization for time series data. Currently, it supports a version with Tikhonov regularization and sparse constraints as proposed by Fabregat R.. et. al. and implemented in Matlab in https://github.com/raimon-fa/palm-nmf. It uses a PALM optimization scheme. We plan to add other models and optimization algorithms.
- This work has been supported by NSF award #1849930 and the Gordon and Betty Moore and Alfred P. Sloan Foundations Data Science Environments grant (MSDSE).