Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Online implementation of Non-negative Matrix Factorizarion (NMF)

See original GitHub issue

Description

@GaelVaroquaux Maybe an online version of the NMF could be useful in big data settings.

Steps/Code to Reproduce

Expected Results

Actual Results

Here I show the results for an sparse matrix as input of size 1M x 2.8k The factorization has n_components = 10 and each point is an entire pass on the data. The current online implementation only supports the kullback-leibler divergence, but in practice can be generalized to any beta divergence.

batch_vs_online_nmf

Versions

Issue Analytics

State:
Created 5 years ago
Reactions:1
Comments:19 (18 by maintainers)

Top GitHub Comments

1reaction

jeremiedbbcommented, May 4, 2020

So in all your benchmarks, online NMF converges faster to a better solution. Pretty convincing !

1reaction

cmarmocommented, May 3, 2020

@jeremiedbb this new plot and my previous comment, are they answering your question? bench_topics

Top Results From Across the Web

Online Nonnegative Matrix Factorization with Outliers - arXiv

Abstract—We propose a unified and systematic framework for performing online nonnegative matrix factorization in the presence of outliers.

Online algorithms for Nonnegative Matrix Factorization with ...

We propose an algorithm to estimate Itakura-. Saito NMF (IS-NMF) on audio signals of possibly infinite duration with tractable memory and time complexity....

Efficient Document Clustering via Online Nonnegative Matrix ...

Abstract. In recent years, Nonnegative Matrix Factorization. (NMF) has received considerable interest from the data mining and information retrieval fields.

Online Projective Nonnegative Matrix Factorization for Large ...

Projective Nonnegative Matrix Factorization (PNMF) is one of the recent methods for computing low-rank approximations to data matrices.

Non-Negative Matrix Factorization - GeeksforGeeks

NMF is used in major applications such as image processing, text mining, spectral data analysis and many more. Currently, there is an ongoing ......