Using scikit-learn NMF with ONE precomputed basis vector

I am using scikit-learn NMF to decompose a matrix I have to 2-4 bases/factors.

I have only 1 factor which I know/want to anchor, namely I know W(1) and I want to use that knowledge in my factorization process. Is that possible?

I know I can use scikit-learn NMF with a fixed H or W matrix (see this question and answer) in case I have the entire matrix fixed. But what if I only have one basis/factor fixed? Is that technically possible to achieve with this function?



from Recent Questions - Stack Overflow https://ift.tt/3777XMP
https://ift.tt/eA8V8J

Comments

Popular posts from this blog

Spring Elasticsearch Operations

Hibernate Search - Elasticsearch with JSON manipulation

Today Walkin 14th-Sept