在线时间:8:00-16:00
迪恩网络APP
随时随地掌握行业动态
扫描二维码
关注迪恩网络微信公众号
开源软件名称(OpenSource Name):kimjingu/nonnegfac-matlab开源软件地址(OpenSource Url):https://github.com/kimjingu/nonnegfac-matlab开源编程语言(OpenSource Language):MATLAB 100.0%开源软件介绍(OpenSource Introduction):Nonnegative Matrix and Tensor Factorization Algorithms ToolboxThis package includes MATLAB implementations of fast optimization algorithms for computing nonnegative matrix and tensor factorizations. Nonnegative Matrix Factorization (NMF)nmf.m is a program for executing NMF algorithms. When A is a nonnegative matrix,
returns the NMF of A with 10 as a target lower-rank. The two parameters (input
data matrix and target lower-rank) are mandatory, whereas other parameters are
optional. An appropriate value for the target lower-rank depends on each data
matrix A and on the purpose of performing NMF. To learn optional parameters,
open nmf.m and see the descriptions there. For example, the default algorithm
for computing NMF, which is
Names of NMF algorithms implemented are as follows.
Several usage examples are provided in example_nmf_1.m. Another example file, example_nmf_2.m, shows how it can be tested whether an NMF algorithm recovers true latent factors when applied to a synthetic matrix whose latent factors are known. Default NMF algorithm is Nonnegative Tensor Factorization (Nonnegative CP)This software performs nonnegative tensor factorization in CP (Canonical Decomposition / PARAFAC) model. ncp.m is a program for executing NTF algorithms. To use this program, it is necessary to first install MATLAB Tensor Toolbox by Brett W. Bader and Tamara G. Kolda, available at http://csmr.ca.sandia.gov/~tgkolda/TensorToolbox/. The latest version that was tested with this program is Version 2.4, March 2010. Refer to the help manual of the toolbox for installation and basic usage. Please see the description in ncp.m and try to execute example_ncp.m to learn how to use this program. Names of nonnegative CP algorithms implemented are as follows.
References
FeedbackPlease send bug reports, comments, or questions to Jingu Kim. Contributions and extentions with new algorithms are welcome. |
2023-10-27
2022-08-15
2022-08-17
2022-09-23
2022-08-13
请发表评论