How does ICA implement n_components? #30701
Unanswered
dherrera1911
asked this question in
Q&A
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
When I call
help(FastICA), the docstring says that the implementation is based on "A. Hyvarinen and E. Oja, Independent Component Analysis: Algorithms and Applications". However, I'm not sure from this reference what is the implementation of the optionn_components. The paper in question only treats the case where number of components equals the number of dimensions, and dimensionality reduction would be done in the pre-processing with PCA. However, in the implementation I also see the following comment:"While FastICA was proposed to estimate as many sources as features, it is possible to estimate less by setting n_components < n_features. It this case K is not a square matrix and the estimated A is the pseudo-inverse of
W K.". Maybe a reference for this should be included in the docstring for FastICA?Beta Was this translation helpful? Give feedback.
All reactions